48 GB · 445 reports

RTX 4090D · 48 GB

Qualified results first; original reports and full tables stay below.
Reset

All tasks · Qualified results

Speeds compare only within one exact test. Recipe availability is checked on each report against the published recipe executor.

No qualified results yet · All tasks · RTX 4090D · 48 GB

445 original reports are kept as evidence but do not qualify for comparison; each report states why. This does not show that the model or GPU cannot run it.

Measured alternatives on other GPUs

Measured on different GPUs with the same or smaller listed capacity. These are references for choosing hardware, not performance or compatibility evidence for your selected GPU.

  1. Qwen3.8-27B-NVFP4

    RadixArk · 2 tested configurations · 1 GPU

    • RTX 5090 · 32 GBNVFP4 · 8K → 2K tokens
      Output throughput · includes prefill · 1 request
      70.8tok/s
      Observed GPU occupancy
      27.3GiB
      First token
      633 ms
      Test details

      Total output tokens divided by the full request window, including prefill and queueing.

      Occupancy includes engine buffers and reserved KV cache. It is not the minimum VRAM required by the model.

      Reports combined counts reports, not inference requests. Request successes and failures are listed in each report.

      Engine
      vllm
      Protocol
      wbench-fixed-http-v1
      Dataset
      synthetic-tokenizer-v1:repeats=2
      Workload
      8192in-2048out
      ConcurrencyMean throughput (tok/s)P95TTFTObserved occupancy (GiB)GPU power (W)Reports combinedView report
      C170.8229.01 s633 ms27.29405.51bad294c0 ↗
    • 1 other test configurations on this GPU
      • NVFP4 · 32K → 2K tokens
        Output throughput · includes prefill · 1 request
        62.5tok/s
        Observed GPU occupancy
        27.3GiB
        First token
        3.6 s
        Test details

        Total output tokens divided by the full request window, including prefill and queueing.

        Occupancy includes engine buffers and reserved KV cache. It is not the minimum VRAM required by the model.

        Reports combined counts reports, not inference requests. Request successes and failures are listed in each report.

        Engine
        vllm
        Protocol
        wbench-fixed-http-v1
        Dataset
        synthetic-tokenizer-v1:repeats=2
        Workload
        32768in-2048out
        ConcurrencyMean throughput (tok/s)P95TTFTObserved occupancy (GiB)GPU power (W)Reports combinedView report
        C162.4832.79 s3.6 s27.29433.31746bbc39 ↗
  2. clef-flash

    Cloudflare · 4 tested configurations · 1 GPU

    • vGPU-32GB · 32 GBbf16 · Text → Structured data · typed-invoices-v1 · Test duration 1 s
      Throughput · Batch 1
      10.3decisions/sec
      Best measured · Batch 4
      10.9decisions/sec
      Observed GPU occupancy
      18.2GiB
      P95 latency
      104 ms
      Test details · 2 batch sizes

      Occupancy includes engine buffers and reserved KV cache. It is not the minimum VRAM required by the model.

      Reports combined counts reports, not inference requests. Request successes and failures are listed in each report.

      Engine
      clef-native
      Protocol
      native-forward-v1
      Dataset
      typed-invoices-v1
      Workload
      decision
      Batch sizeMean throughput (decisions/sec)P95TTFTObserved occupancy (GiB)GPU power (W)Reports combinedView report
      B110.26104 ms—18.1767.911d1661ad ↗
      B410.89389 ms—18.4915611d1661ad ↗
    • 3 other test configurations on this GPU
      • bf16 · Text → Structured data · typed-invoices-v1 · Test duration 6 s
        Throughput · Batch 1
        10.6decisions/sec
        Best measured · Batch 4
        11decisions/sec
        Observed GPU occupancy
        18.2GiB
        P95 latency
        99 ms
        Test details · 2 batch sizes

        Occupancy includes engine buffers and reserved KV cache. It is not the minimum VRAM required by the model.

        Reports combined counts reports, not inference requests. Request successes and failures are listed in each report.

        Engine
        clef-native · PyTorch 2.13.0+cu130
        Protocol
        native-forward-v1
        Dataset
        typed-invoices-v1
        Workload
        decision
        Batch sizeMean throughput (decisions/sec)P95TTFTObserved occupancy (GiB)GPU power (W)Reports combinedView report
        B110.5699 ms—18.17206.1263f44036 ↗
        B411.04402 ms—18.55244.4263f44036 ↗
      • bf16 · Text → Structured data · typed-invoices-v1 · Test duration 6 s
        Throughput · Batch 1
        10.6decisions/sec
        Best measured · Batch 4
        11.1decisions/sec
        Observed GPU occupancy
        18.2GiB
        P95 latency
        103 ms
        Test details · 2 batch sizes

        Occupancy includes engine buffers and reserved KV cache. It is not the minimum VRAM required by the model.

        Reports combined counts reports, not inference requests. Request successes and failures are listed in each report.

        Engine
        clef-native
        Protocol
        native-forward-v1
        Dataset
        typed-invoices-v1
        Workload
        decision
        Batch sizeMean throughput (decisions/sec)P95TTFTObserved occupancy (GiB)GPU power (W)Reports combinedView report
        B110.56103 ms—18.17196.814eb73a4d ↗
        B411.1390 ms—18.56235.914eb73a4d ↗
      • bf16 · 1K → 1K tokens
        Throughput · Batch 1
        10.2decisions/sec
        Best measured · Batch 4
        10.9decisions/sec
        Observed GPU occupancy
        18.2GiB
        P95 latency
        104 ms
        Test details · 2 batch sizes

        Occupancy includes engine buffers and reserved KV cache. It is not the minimum VRAM required by the model.

        Reports combined counts reports, not inference requests. Request successes and failures are listed in each report.

        Engine
        vLLM
        Protocol
        native-forward-v1
        Dataset
        typed-invoices-v1
        Workload
        decision
        Batch sizeMean throughput (decisions/sec)P95TTFTObserved occupancy (GiB)GPU power (W)Reports combinedView report
        B110.19104 ms—18.1769.81caef7154 ↗
        B410.88388 ms—18.49152.81caef7154 ↗

Occupancy includes engine buffers and reserved KV cache. It is not the minimum VRAM required by the model.

Quality by dataset

Scores from different datasets and sample scopes are not directly comparable.
Quality by dataset · 7 rows
ModelDatasetFormatEngineAverage scoreSample sizeRuns
RadixArk/Qwen3.8-27B-NVFP4hle-test-self-judge-22ed307-v1-n5-mt16384-t0NVFP4vllm20%5 / 2500 · Subset only
1 runs20% · 7bbadab5 ↗
RadixArk/Qwen3.8-27B-NVFP4simpleqa-verified-eval-self-judge-0dc97e0-v1-n1000-mt16384-t0NVFP4vllm6.4%1000 / 1000 · Full dataset
1 runs6.4% · 404eb015 ↗
RadixArk/Qwen3.8-27B-NVFP4simpleqa-verified-eval-self-judge-0dc97e0-v1-n5-mt16384-t0NVFP4vllm0%5 / 1000 · Subset only
1 runs0% · 328fd742 ↗
RadixArk/Qwen3.8-27B-NVFP4ifbench-test-official-loose-1c40f0c-v1-n300-mt16384-t0NVFP4vllm67.67%300 / 300 · Full dataset
1 runs67.67% · 02a43d63 ↗
RadixArk/Qwen3.8-27B-NVFP4ifbench-test-official-loose-1c40f0c-v1-n3-mt16384-t0NVFP4vllm66.67%3 / 300 · Subset only
1 runs66.67% · 20ef4b9f ↗
RadixArk/Qwen3.8-27B-NVFP4aime2026-test-zero-shot-strict-v4-n30-mt16384-t0NVFP4vllm63.33%30 / 30 · Full dataset
1 runs63.33% · 87acbd23 ↗
RadixArk/Qwen3.8-27B-NVFP4mmlu-pro-test-zero-shot-strict-v4-n3-mt16384-t0NVFP4vllm66.67%3 / 12032 · Subset only
1 runs66.67% · 1d2bdd31 ↗
Measurements without a comparable protocol · 917 rows

These measured rates remain visible, but are excluded from comparison curves because the protocol is unknown.

Measurements without a comparable protocol · 917 rows
ModelFormatEngineWorkloadConcurrencyThroughputGPU power (W)
Gemma-3-270m ↗BF16vLLMMixed1558.43 tok/s185.07
Gemma-3-270m ↗BF16vLLMMixed2861.15 tok/s164.61
Gemma-3-270m ↗BF16vLLMMixed41,559.22 tok/s178.04
Gemma-3-270m ↗BF16vLLMMixed82,479.12 tok/s161.51
Gemma-3-270m ↗BF16vLLM32K1532.47 tok/s189.12
Gemma-3-270m ↗BF16vLLM32K2827.48 tok/s173.12
Gemma-3-270m ↗BF16vLLM32K41,522.33 tok/s192.63
Gemma-3-270m ↗BF16vLLM32K81,458.27 tok/s157.46
Gemma-3-270m ↗BF16vLLM16K1546.4 tok/s186.78
Gemma-3-270m ↗BF16vLLM16K2856.9 tok/s169.84
Gemma-3-270m ↗BF16vLLM16K41,630.3 tok/s187.19
Gemma-3-270m ↗BF16vLLM16K81,552.35 tok/s149.86
Gemma-3-270m ↗BF16vLLM8K1553.45 tok/s184.86
Gemma-3-270m ↗BF16vLLM8K81,651.81 tok/s146.62
Gemma-3-270m ↗BF16vLLM8K325,116.35 tok/s192.84
Gemma-3-270m ↗BF16vLLM8K1288,584.2 tok/s252.28
Gemma-3-270m ↗BF16vLLM8K5128,452.13 tok/s267.55
Gemma-3-270m ↗BF16vLLM4K1566.15 tok/s185.97
Gemma-3-270m ↗BF16vLLM4K81,602.47 tok/s142.08
Gemma-3-270m ↗BF16vLLM4K325,431.31 tok/s184.86
Gemma-3-270m ↗BF16vLLM4K1288,967.4 tok/s240.86
Gemma-3-270m ↗BF16vLLM4K5129,211.88 tok/s261.53
Gemma-3-270m ↗BF16vLLM2K1566.41 tok/s184.31
Gemma-3-270m ↗BF16vLLM2K81,649.66 tok/s141.18
Gemma-3-270m ↗BF16vLLM2K325,509.22 tok/s179.5
Gemma-3-270m ↗BF16vLLM2K1289,516.01 tok/s234.75
Gemma-3-270m ↗BF16vLLM2K5129,785.93 tok/s260.17
Gemma-3-270m ↗BF16vLLM1K1566.96 tok/s184.81
Gemma-3-270m ↗BF16vLLM1K81,687.01 tok/s140.83
Gemma-3-270m ↗BF16vLLM1K325,586.19 tok/s178.08
Gemma-3-270m ↗BF16vLLM1K1289,836.61 tok/s232.35
Gemma-3-270m ↗BF16vLLM1K51210,434.61 tok/s262.5
Gemma-3-270m ↗BF16vLLM5121598.27 tok/s189.67
Gemma-3-270m ↗BF16vLLM51281,695.08 tok/s138.95
Gemma-3-270m ↗BF16vLLM512325,829.92 tok/s171.14
Gemma-3-270m ↗BF16vLLM51212810,407.53 tok/s218.17
Gemma-3-270m ↗BF16vLLM51251211,978.52 tok/s250.64
Gemma-3-270m ↗BF16vLLM2561640.38 tok/s195.23
Gemma-3-270m ↗BF16vLLM25681,917.24 tok/s142.39
Gemma-3-270m ↗BF16vLLM256326,067.09 tok/s161.77
Gemma-3-270m ↗BF16vLLM25612812,576.67 tok/s211.73
Gemma-3-270m ↗BF16vLLM25651214,801.51 tok/s230.89
Gemma-3-270m ↗BF16vLLM1281682.05 tok/s200.71
Gemma-3-270m ↗BF16vLLM12882,330.32 tok/s153.08
Gemma-3-270m ↗BF16vLLM128326,305 tok/s155.19
Gemma-3-270m ↗BF16vLLM12812814,741.29 tok/s209.43
Gemma-3-270m ↗BF16vLLM1281648.15 tok/s196.02
Gemma-3-270m ↗BF16vLLM12882,667.02 tok/s161.76
Gemma-3-270m ↗BF16vLLM128326,590.82 tok/s162.13
Gemma-3-270m ↗BF16vLLM12812814,844.36 tok/s218.56
Gemma-3-270m ↗BF16vLLM12851218,775.66 tok/s242.81
Gemma-3-270m ↗BF16vLLMMixed1327.11 tok/s173.29
Gemma-3-270m ↗BF16vLLMMixed2358.96 tok/s184.1
Gemma-3-270m ↗BF16vLLMMixed4431.84 tok/s235.46
Gemma-3-270m ↗BF16vLLMMixed8331.25 tok/s228.1
Gemma-3-270m ↗BF16vLLM4K1447.36 tok/s174.14
Gemma-3-270m ↗BF16vLLM4K81,552.03 tok/s149.96
Gemma-3-270m ↗BF16vLLM4K324,736.15 tok/s188.5
Gemma-3-270m ↗BF16vLLM4K1287,607.28 tok/s243.69
Gemma-3-270m ↗BF16vLLM4K5125,229.81 tok/s252.58
Gemma-3-270m ↗BF16vLLM2K1562.52 tok/s185.58
Gemma-3-270m ↗BF16vLLM2K325,235.34 tok/s181.53
Gemma-3-270m ↗BF16vLLM2K1288,913.21 tok/s243.45
Gemma-3-270m ↗BF16vLLM2K5127,885.21 tok/s266.09
Gemma-3-270m ↗BF16vLLM32K1536.84 tok/s190.17
Gemma-3-270m ↗BF16vLLM16K1260.85 tok/s152.74
Gemma-3-270m ↗BF16vLLM1K81,599.27 tok/s149.04
Gemma-3-270m ↗BF16vLLM1K325,070.66 tok/s185.03
Gemma-3-270m ↗BF16vLLM1K1289,108.9 tok/s248.31
Gemma-3-270m ↗BF16vLLM512137.2 tok/s269.12
Gemma-3-270m ↗BF16vLLM51281,339.85 tok/s185.83
Gemma-3-270m ↗BF16vLLM512325,806.54 tok/s172.88
Gemma-3-270m ↗BF16vLLM51212810,299.83 tok/s228.27
Gemma-3-270m ↗BF16vLLM5125128,939.01 tok/s261.82
Gemma-3-270m ↗BF16vLLM2K1436.43 tok/s170.53
Gemma-3-270m ↗BF16vLLM2K81,445.3 tok/s164.85
Gemma-3-270m ↗BF16vLLM2K325,480.34 tok/s181.78
Gemma-3-270m ↗BF16vLLM2K1286,609.37 tok/s263.43
Gemma-3-270m ↗BF16vLLM2K5129,695.96 tok/s263.99
Gemma-3-270m ↗BF16vLLM25681,632.06 tok/s141.7
Gemma-3-270m ↗BF16vLLM256325,676.02 tok/s176.12
Gemma-3-270m ↗BF16vLLM2561286,102.07 tok/s253.16
Gemma-3-270m ↗BF16vLLM25651214,312.31 tok/s236.52
Gemma-3-270m ↗BF16vLLM1K1565.84 tok/s184.72
Gemma-3-270m ↗BF16vLLM1K81,665.74 tok/s141.35
Gemma-3-270m ↗BF16vLLM1K325,519.73 tok/s177.48
Gemma-3-270m ↗BF16vLLM1K1289,781.99 tok/s237.32
Gemma-3-270m ↗BF16vLLM1K51210,115.03 tok/s263.25
Gemma-3-270m ↗BF16vLLM5121608.88 tok/s191.83
Gemma-3-270m ↗BF16vLLM51281,713.17 tok/s140.12
Gemma-3-270m ↗BF16vLLM512322,851.12 tok/s205.79
Gemma-3-270m ↗BF16vLLM51212810,816.18 tok/s219.66
Gemma-3-270m ↗BF16vLLM51251212,144.95 tok/s252.42
Gemma-3-270m ↗BF16vLLM1281267.49 tok/s146.51
Gemma-3-270m ↗BF16vLLM12881,699.55 tok/s158.78
Gemma-3-270m ↗BF16vLLM128326,225.67 tok/s157.92
Gemma-3-270m ↗BF16vLLM12812812,831.27 tok/s199.11
Gemma-3-270m ↗BF16vLLM12851214,492.99 tok/s231.59
Gemma-3-270m ↗BF16vLLM25681,747.17 tok/s143.73
Gemma-3-270m ↗BF16vLLM256326,119.2 tok/s169.44
Gemma-3-270m ↗BF16vLLM25612812,219.3 tok/s205.72
Gemma-3-270m ↗BF16vLLM25651214,354.97 tok/s230.05
Gemma-3-270m ↗BF16vLLM1281681.68 tok/s200.79
Gemma-3-270m ↗BF16vLLM12881,846.76 tok/s138.58
Gemma-3-270m ↗BF16vLLM128326,287.08 tok/s156.81
Gemma-3-270m ↗BF16vLLM1281692.79 tok/s202.09
Gemma-3-270m ↗BF16vLLM12882,225.08 tok/s146.97
Gemma-3-270m ↗BF16vLLM128326,494.38 tok/s156.56
Gemma-3-270m ↗BF16vLLM12812815,588.69 tok/s212.36
Gemma-3-270m ↗BF16vLLM12851218,723.26 tok/s227.31
Hunyuan-1.8B-FP8 ↗FP8vLLM2K1249.24 tok/s211.56
Hunyuan-1.8B-FP8 ↗FP8vLLM2K81,534.73 tok/s214.07
Hunyuan-1.8B-FP8 ↗FP8vLLM2K162,695.85 tok/s231.18
Hunyuan-1.8B-FP8 ↗FP8vLLM2K323,687.18 tok/s217.51
Hunyuan-1.8B-FP8 ↗FP8vLLM2K645,035.17 tok/s235.97
Hunyuan-1.8B-FP8 ↗FP8vLLM2K1286,043.67 tok/s254.45
Hunyuan-1.8B-FP8 ↗FP8vLLM2K2566,554.34 tok/s271.02
Hunyuan-1.8B-FP8 ↗FP8vLLM2K5126,777 tok/s271.56
Hunyuan-1.8B-FP8 ↗FP8vLLM2K1249.26 tok/s209.3
Hunyuan-1.8B-FP8 ↗FP8vLLM2K81,547.49 tok/s212.55
Hunyuan-1.8B-FP8 ↗FP8vLLM2K162,706.58 tok/s228.46
Hunyuan-1.8B-FP8 ↗FP8vLLM2K323,680.94 tok/s216.53
Hunyuan-1.8B-FP8 ↗FP8vLLM2K645,049.38 tok/s234.03
Hunyuan-1.8B-FP8 ↗FP8vLLM2K1286,086.99 tok/s252.24
Hunyuan-1.8B-FP8 ↗FP8vLLM2K2566,593.06 tok/s270.02
Hunyuan-1.8B-FP8 ↗FP8vLLM2K5126,772.9 tok/s269.82
Hunyuan-1.8B-FP8 ↗FP8vLLM2K1249.58 tok/s208.11
Hunyuan-1.8B-FP8 ↗FP8vLLM2K81,527.27 tok/s211.02
Hunyuan-1.8B-FP8 ↗FP8vLLM2K162,671.61 tok/s226.59
Hunyuan-1.8B-FP8 ↗FP8vLLM2K323,648.6 tok/s214.73
Hunyuan-1.8B-FP8 ↗FP8vLLM2K645,112.76 tok/s231.93
Hunyuan-1.8B-FP8 ↗FP8vLLM2K1286,017.48 tok/s250.6
Hunyuan-1.8B-FP8 ↗FP8vLLM2K2566,523.32 tok/s270.23
Hunyuan-7B-FP8 ↗FP8vLLM2K185.79 tok/s263.43
Hunyuan-7B-FP8 ↗FP8vLLM2K8591.72 tok/s258.46
Hunyuan-7B-FP8 ↗FP8vLLM2K16968.24 tok/s252.7
Hunyuan-7B-FP8 ↗FP8vLLM2K321,709.98 tok/s266.64
Hunyuan-7B-FP8 ↗FP8vLLM2K642,654.38 tok/s293.61
Hunyuan-7B-FP8 ↗FP8vLLM2K1283,096.26 tok/s312.51
Hunyuan-7B-FP8 ↗FP8vLLM2K2564,010.29 tok/s308.72
Hunyuan-7B-FP8 ↗FP8vLLM2K5124,024.58 tok/s309.83
Hunyuan-7B-FP8 ↗FP8vLLM2K185.71 tok/s262.3
Hunyuan-7B-FP8 ↗FP8vLLM2K8591.65 tok/s257.54
Hunyuan-7B-FP8 ↗FP8vLLM2K16866.94 tok/s252.35
Hunyuan-7B-FP8 ↗FP8vLLM2K321,682.08 tok/s266.2
Hunyuan-7B-FP8 ↗FP8vLLM2K642,668.28 tok/s295.58
Hunyuan-7B-FP8 ↗FP8vLLM2K185.9 tok/s257.47
Hunyuan-7B-FP8 ↗FP8vLLM2K8582.69 tok/s254.22
Hunyuan-7B-FP8 ↗FP8vLLM2K16972.07 tok/s247.73
Hunyuan-7B-FP8 ↗FP8vLLM2K321,687.45 tok/s264.22
Hunyuan-7B-FP8 ↗FP8vLLM2K642,690.73 tok/s292.87
Hunyuan-7B-FP8 ↗FP8vLLM2K1283,515.54 tok/s312.46
Hunyuan-7B-FP8 ↗FP8vLLM2K2563,833.43 tok/s306.58
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K1420.66 tok/s246.98
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K2526.72 tok/s211.09
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K4903.58 tok/s226.18
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K81,205.87 tok/s249.31
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K161,221.5 tok/s249.53
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K1424.64 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K2534.4 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K4899.87 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K81,216.08 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K161,218.67 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K1426.5 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K2536.37 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K4901.77 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K81,219.05 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K161,192.23 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K1424.7 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K2532.7 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K4902.61 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K81,205.36 tok/s—
HY-MT1.5-1.8B-Q4_K_M ↗Q4_K_MvLLM1K161,201.19 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K1364.04 tok/s279.69
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K2489.7 tok/s229.33
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K4872.37 tok/s243.84
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K81,206.15 tok/s258.49
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K161,221.48 tok/s256.97
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K1367.11 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K2494.42 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K4871 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K81,215.96 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K161,220.63 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K1368.57 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K2495.99 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K4861.42 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K81,138.09 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K161,220.29 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K1367.88 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K2494.33 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K4855.11 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K81,219.77 tok/s—
HY-MT1.5-1.8B-Q6_K ↗Q6_KvLLM1K161,221.21 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K1313.55 tok/s243.16
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K2439.37 tok/s206.75
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K4803.48 tok/s218.36
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K81,236.89 tok/s238.01
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K161,240.99 tok/s241.67
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K1317.91 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K2447.85 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K4813.28 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K81,229.45 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K161,241.06 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K1318.22 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K2445.88 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K4810.27 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K81,180.43 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K161,186.06 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K1320.03 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K2447.2 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K4815.93 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K81,213.14 tok/s—
HY-MT1.5-1.8B-Q8_0 ↗Q8_0vLLM1K161,179.37 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K1165.14 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K2265.72 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K4423.84 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K8578.74 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K16607.92 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K1165.13 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K2259.11 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K4418.02 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K8552.39 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K16581.37 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K1164.97 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K2262.95 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K4456.01 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K8533.33 tok/s—
HY-MT1.5-7B-Q4_K_M ↗Q4_K_MvLLM1K16544.52 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K1130.45 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K2219.72 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K4389.11 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K8580.67 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K16595.63 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K1130.28 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K2216.5 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K4369.93 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K8553.33 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K16519.12 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K1130.15 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K2213.5 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K4383.63 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K8570.85 tok/s—
HY-MT1.5-7B-Q6_K ↗Q6_KvLLM1K16539.62 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K1105.45 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K2182.38 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K4326.41 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K8582.7 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K16570.76 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K1105.42 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K2179.71 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K4339.75 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K8542.49 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K16510.31 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K1105.18 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K2180.67 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K4335.84 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K8487.05 tok/s—
HY-MT1.5-7B-Q8_0 ↗Q8_0vLLM1K16499.3 tok/s—
Qwen2.5-0.5B-Q4_K_M ↗Q4_K_MvLLM1K1749.41 tok/s—
Qwen2.5-0.5B-Q4_K_M ↗Q4_K_MvLLM1K2794.85 tok/s—
Qwen2.5-0.5B-Q4_K_M ↗Q4_K_MvLLM1K4922.16 tok/s—
Qwen2.5-0.5B-Q4_K_M ↗Q4_K_MvLLM1K81,055 tok/s—
Qwen2.5-0.5B-Q4_K_M ↗Q4_K_MvLLM1K161,378.74 tok/s—
Qwen2.5-72B-AWQ ↗AWQvLLM2K122.47 tok/s335.22
Qwen2.5-72B-AWQ ↗AWQvLLM2K8159.82 tok/s331.5
Qwen2.5-72B-AWQ ↗AWQvLLM2K16290.43 tok/s356.86
Qwen2.5-72B-AWQ ↗AWQvLLM2K32492.31 tok/s391.28
Qwen2.5-72B-AWQ ↗AWQvLLM2K64568.3 tok/s385.25
Qwen2.5-72B-AWQ ↗AWQvLLM2K128646.86 tok/s398.54
Qwen2.5-72B-AWQ ↗AWQvLLM2K256594.19 tok/s392.55
Qwen2.5-72B-AWQ ↗AWQvLLM2K512597.36 tok/s397.72
Qwen2.5-72B-AWQ ↗AWQvLLM2K122.51 tok/s329.21
Qwen2.5-72B-AWQ ↗AWQvLLM2K8156.01 tok/s329.45
Qwen2.5-72B-AWQ ↗AWQvLLM2K16285.03 tok/s353.6
Qwen2.5-72B-AWQ ↗AWQvLLM2K32267.6 tok/s381.6
Qwen2.5-72B-AWQ ↗AWQvLLM2K64583.13 tok/s382.87
Qwen2.5-72B-AWQ ↗AWQvLLM2K128642.1 tok/s393.7
Qwen2.5-72B-AWQ ↗AWQvLLM2K256591.39 tok/s386.76
Qwen3-0.6B ↗BF16vLLM2K1368.05 tok/s220.2
Qwen3-0.6B ↗BF16vLLM2K81,925.32 tok/s222.84
Qwen3-0.6B ↗BF16vLLM2K162,677.74 tok/s201.02
Qwen3-0.6B ↗BF16vLLM2K324,179.65 tok/s234.44
Qwen3-0.6B ↗BF16vLLM2K645,676.24 tok/s261.53
Qwen3-0.6B ↗BF16vLLM2K1286,268.94 tok/s279.97
Qwen3-0.6B ↗BF16vLLM2K2566,682.18 tok/s290.79
Qwen3-0.6B ↗BF16vLLM2K5126,633 tok/s289.9
Qwen3-0.6B-FP8 ↗FP8vLLM2K1229.63 tok/s146.12
Qwen3-0.6B-FP8 ↗FP8vLLM2K81,598.01 tok/s174.09
Qwen3-0.6B-FP8 ↗FP8vLLM2K162,500.48 tok/s163.99
Qwen3-0.6B-FP8 ↗FP8vLLM2K324,118.01 tok/s187.59
Qwen3-0.6B-FP8 ↗FP8vLLM2K646,241.98 tok/s219.05
Qwen3-0.6B-FP8 ↗FP8vLLM2K1288,088.78 tok/s262.62
Qwen3-0.6B-FP8 ↗FP8vLLM2K2569,378.78 tok/s286.19
Qwen3-0.6B-FP8 ↗FP8vLLM2K5129,322.34 tok/s289.15
Qwen3-0.6B-FP8 ↗FP8vLLM2K10249,270.05 tok/s290.67
Qwen3-0.6B-FP8 ↗FP8vLLM2K1228.61 tok/s145.63
Qwen3-0.6B-FP8 ↗FP8vLLM2K81,580.25 tok/s171.1
Qwen3-0.6B-FP8 ↗FP8vLLM2K162,485.19 tok/s163.23
Qwen3-0.6B-FP8 ↗FP8vLLM2K324,092.97 tok/s186.86
Qwen3-0.6B-FP8 ↗FP8vLLM2K646,246.15 tok/s218.29
Qwen3-0.6B-FP8 ↗FP8vLLM2K1288,111.09 tok/s260.13
Qwen3-0.6B-FP8 ↗FP8vLLM2K2569,389.58 tok/s287.53
Qwen3-0.6B-FP8 ↗FP8vLLM2K5129,314.63 tok/s288.79
Qwen3-0.6B-FP8 ↗FP8vLLM2K10249,293.75 tok/s288.38
Qwen3-0.6B-FP8 ↗FP8vLLM2K1230.12 tok/s143.62
Qwen3-0.6B-FP8 ↗FP8vLLM2K81,590.87 tok/s170.77
Qwen3-0.6B-FP8 ↗FP8vLLM2K162,490.38 tok/s161.21
Qwen3-0.6B-FP8 ↗FP8vLLM2K324,094.54 tok/s182.77
Qwen3-0.6B-FP8 ↗FP8vLLM2K646,224.49 tok/s214.96
Qwen3-0.6B-FP8 ↗FP8vLLM2K1288,090.46 tok/s255.69
Qwen3-0.6B-FP8 ↗FP8vLLM2K2569,380.9 tok/s283.27
Qwen3-0.6B-FP8 ↗FP8vLLM2K5129,328.49 tok/s285.28
Qwen3-0.6B-FP8 ↗FP8vLLM2K1231.11 tok/s144.22
Qwen3-0.6B-FP8 ↗FP8vLLM2K81,592.69 tok/s171.19
Qwen3-0.6B-FP8 ↗FP8vLLM2K162,491.82 tok/s161.07
Qwen3-0.6B-FP8 ↗FP8vLLM2K324,090.68 tok/s183.52
Qwen3-0.6B-FP8 ↗FP8vLLM2K646,245.89 tok/s216.88
Qwen3-0.6B-FP8 ↗FP8vLLM2K1288,087.53 tok/s256.21
Qwen3-0.6B-FP8 ↗FP8vLLM2K2569,373.14 tok/s281.63
Qwen3-0.6B-FP8 ↗FP8vLLM2K1230.43 tok/s140.25
Qwen3-0.6B-FP8 ↗FP8vLLM2K81,591.62 tok/s171.52
Qwen3-0.6B-FP8 ↗FP8vLLM2K162,485.11 tok/s162.18
Qwen3-0.6B-FP8 ↗FP8vLLM2K324,088.45 tok/s185.66
Qwen3-0.6B-FP8 ↗FP8vLLM2K646,255.71 tok/s218.61
Qwen3-0.6B-FP8 ↗FP8vLLM2K1288,109.16 tok/s260.79
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K1565.03 tok/s186.89
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K2665.75 tok/s150.73
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K41,105.47 tok/s157.62
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K81,490.92 tok/s169.76
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K161,490.8 tok/s170.77
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K1565.23 tok/s190.9
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K2665.32 tok/s155.08
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K41,103.61 tok/s160.07
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K81,456.73 tok/s172.69
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K161,478.49 tok/s170.53
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K1571.89 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K2665.58 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K41,100.18 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K81,470.9 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K161,471.65 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K1569.89 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K2667.91 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K41,104.22 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K81,481.7 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K161,483.75 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K1569.79 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K2667.55 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K41,099.69 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K81,478.81 tok/s—
Qwen3-0.6B-Q8_0 ↗Q8_0vLLM1K161,477.03 tok/s—
Qwen3-1.7B ↗BF16vLLM2K1196.71 tok/s278.09
Qwen3-1.7B ↗BF16vLLM2K81,202.43 tok/s250.26
Qwen3-1.7B ↗BF16vLLM2K161,931.19 tok/s236.68
Qwen3-1.7B ↗BF16vLLM2K323,118 tok/s263.79
Qwen3-1.7B ↗BF16vLLM2K644,571.17 tok/s276.8
Qwen3-1.7B ↗BF16vLLM2K1285,461.19 tok/s303.63
Qwen3-1.7B ↗BF16vLLM2K2566,019.91 tok/s319.95
Qwen3-1.7B ↗BF16vLLM2K5125,993.48 tok/s319.17
Qwen3-1.7B-FP8 ↗FP8vLLM2K1138.43 tok/s184.6
Qwen3-1.7B-FP8 ↗FP8vLLM2K81,058.96 tok/s201.86
Qwen3-1.7B-FP8 ↗FP8vLLM2K161,750.12 tok/s196.14
Qwen3-1.7B-FP8 ↗FP8vLLM2K322,896.36 tok/s214.16
Qwen3-1.7B-FP8 ↗FP8vLLM2K644,280.04 tok/s242.05
Qwen3-1.7B-FP8 ↗FP8vLLM2K1285,315.85 tok/s268.63
Qwen3-1.7B-FP8 ↗FP8vLLM2K2565,987.98 tok/s282.56
Qwen3-1.7B-FP8 ↗FP8vLLM2K5125,969.07 tok/s285.44
Qwen3-1.7B-FP8 ↗FP8vLLM2K1138.45 tok/s182.55
Qwen3-1.7B-FP8 ↗FP8vLLM2K81,060.97 tok/s201.32
Qwen3-1.7B-FP8 ↗FP8vLLM2K161,752.59 tok/s192.7
Qwen3-1.7B-FP8 ↗FP8vLLM2K322,903.83 tok/s212.69
Qwen3-1.7B-FP8 ↗FP8vLLM2K644,281.55 tok/s238.65
Qwen3-1.7B-FP8 ↗FP8vLLM2K1285,320.59 tok/s265.27
Qwen3-1.7B-FP8 ↗FP8vLLM2K2565,995.28 tok/s282.88
Qwen3-1.7B-FP8 ↗FP8vLLM2K5125,973.7 tok/s280.9
Qwen3-1.7B-FP8 ↗FP8vLLM2K1136.85 tok/s177.93
Qwen3-1.7B-FP8 ↗FP8vLLM2K81,051.26 tok/s196.45
Qwen3-1.7B-FP8 ↗FP8vLLM2K161,739.89 tok/s190.26
Qwen3-1.7B-FP8 ↗FP8vLLM2K322,899.87 tok/s209.93
Qwen3-1.7B-FP8 ↗FP8vLLM2K644,282.26 tok/s236.48
Qwen3-1.7B-FP8 ↗FP8vLLM2K1285,317.53 tok/s264.7
Qwen3-1.7B-FP8 ↗FP8vLLM2K2565,618.03 tok/s278.51
Qwen3-1.7B-FP8 ↗FP8vLLM2K5125,783.67 tok/s280.65
Qwen3-1.7B-FP8 ↗FP8vLLM2K1136.96 tok/s177.68
Qwen3-1.7B-FP8 ↗FP8vLLM2K81,050.25 tok/s196.18
Qwen3-1.7B-FP8 ↗FP8vLLM2K161,738.4 tok/s189.7
Qwen3-1.7B-FP8 ↗FP8vLLM2K322,679.18 tok/s209.15
Qwen3-1.7B-FP8 ↗FP8vLLM2K644,281.05 tok/s236.55
Qwen3-1.7B-FP8 ↗FP8vLLM2K1285,318.32 tok/s265.05
Qwen3-1.7B-FP8 ↗FP8vLLM2K2565,992.59 tok/s279.28
Qwen3-1.7B-FP8 ↗FP8vLLM2K1136.82 tok/s173.65
Qwen3-1.7B-FP8 ↗FP8vLLM2K81,051.67 tok/s199.69
Qwen3-1.7B-FP8 ↗FP8vLLM2K161,738.28 tok/s191.06
Qwen3-1.7B-FP8 ↗FP8vLLM2K322,903.95 tok/s213.06
Qwen3-1.7B-FP8 ↗FP8vLLM2K644,280.71 tok/s239.22
Qwen3-1.7B-FP8 ↗FP8vLLM2K1285,319.29 tok/s264.76
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K1326.17 tok/s240.98
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K2464.24 tok/s208.5
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K4803.71 tok/s216.61
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K81,193.6 tok/s233.56
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K161,194.11 tok/s237.08
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K1331.88 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K2466.66 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K4810.48 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K81,187.4 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K161,185.95 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K1332.24 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K2466.5 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K4811.46 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K81,183.28 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K161,186.92 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K1332.49 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K2466.26 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K4813.68 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K81,186.78 tok/s—
Qwen3-1.7B-Q8_0 ↗Q8_0vLLM1K161,187.61 tok/s—
Qwen3-4B ↗BF16vLLM2K199.14 tok/s302.52
Qwen3-4B ↗BF16vLLM2K8654.49 tok/s269.69
Qwen3-4B ↗BF16vLLM2K161,107.11 tok/s263.53
Qwen3-4B ↗BF16vLLM2K321,618.13 tok/s281.54
Qwen3-4B ↗BF16vLLM2K642,959.2 tok/s298.78
Qwen3-4B ↗BF16vLLM2K1283,097.45 tok/s343.47
Qwen3-4B ↗BF16vLLM2K2563,982.47 tok/s347.41
Qwen3-4B ↗BF16vLLM2K5124,144.53 tok/s348.51
Qwen3-4B-AWQ ↗AWQvLLM2K1209.72 tok/s253.3
Qwen3-4B-AWQ ↗AWQvLLM2K81,272.44 tok/s253.07
Qwen3-4B-AWQ ↗AWQvLLM2K161,903.17 tok/s244.57
Qwen3-4B-AWQ ↗AWQvLLM2K322,922.56 tok/s272.3
Qwen3-4B-AWQ ↗AWQvLLM2K643,470.2 tok/s291.51
Qwen3-4B-AWQ ↗AWQvLLM2K1284,099.79 tok/s304.82
Qwen3-4B-AWQ ↗AWQvLLM2K2564,223.48 tok/s302.05
Qwen3-4B-FP8 ↗FP8vLLM2K174.14 tok/s200.2
Qwen3-4B-FP8 ↗FP8vLLM2K8612.17 tok/s216.25
Qwen3-4B-FP8 ↗FP8vLLM2K161,063.11 tok/s212.6
Qwen3-4B-FP8 ↗FP8vLLM2K321,556.33 tok/s223.4
Qwen3-4B-FP8 ↗FP8vLLM2K642,669.39 tok/s242.11
Qwen3-4B-FP8 ↗FP8vLLM2K1283,292.09 tok/s280.22
Qwen3-4B-FP8 ↗FP8vLLM2K2564,015.68 tok/s298.33
Qwen3-4B-FP8 ↗FP8vLLM2K5124,223.99 tok/s298.36
Qwen3-4B-FP8 ↗FP8vLLM2K174.22 tok/s198.86
Qwen3-4B-FP8 ↗FP8vLLM2K8612.74 tok/s215.25
Qwen3-4B-FP8 ↗FP8vLLM2K16847.59 tok/s210.24
Qwen3-4B-FP8 ↗FP8vLLM2K321,535.83 tok/s223.71
Qwen3-4B-FP8 ↗FP8vLLM2K642,885.55 tok/s240.16
Qwen3-4B-FP8 ↗FP8vLLM2K1283,049.61 tok/s277.57
Qwen3-4B-FP8 ↗FP8vLLM2K2564,039.04 tok/s294.68
Qwen3-4B-FP8 ↗FP8vLLM2K5124,248.29 tok/s297.86
Qwen3-4B-FP8 ↗FP8vLLM2K174.15 tok/s196.81
Qwen3-4B-FP8 ↗FP8vLLM2K8607.45 tok/s212.95
Qwen3-4B-FP8 ↗FP8vLLM2K16865.73 tok/s209.24
Qwen3-4B-FP8 ↗FP8vLLM2K321,574.68 tok/s221.34
Qwen3-4B-FP8 ↗FP8vLLM2K642,690.33 tok/s238.98
Qwen3-4B-FP8 ↗FP8vLLM2K1283,249.3 tok/s278.92
Qwen3-4B-FP8 ↗FP8vLLM2K2564,510.88 tok/s292.8
Qwen3-4B-FP8 ↗FP8vLLM2K5124,279.09 tok/s293.24
Qwen3-4B-FP8 ↗FP8vLLM2K173.4 tok/s194.95
Qwen3-4B-FP8 ↗FP8vLLM2K8607.43 tok/s211.72
Qwen3-4B-FP8 ↗FP8vLLM2K16901.53 tok/s206.79
Qwen3-4B-FP8 ↗FP8vLLM2K321,828.23 tok/s219.23
Qwen3-4B-FP8 ↗FP8vLLM2K642,888.89 tok/s239.39
Qwen3-4B-FP8 ↗FP8vLLM2K1283,210.93 tok/s276.06
Qwen3-4B-FP8 ↗FP8vLLM2K2563,957.84 tok/s293.33
Qwen3-4B-FP8 ↗FP8vLLM2K174.3 tok/s192.1
Qwen3-4B-FP8 ↗FP8vLLM2K8599.9 tok/s216.27
Qwen3-4B-FP8 ↗FP8vLLM2K16903.49 tok/s212.2
Qwen3-4B-FP8 ↗FP8vLLM2K321,847.98 tok/s224.12
Qwen3-4B-FP8 ↗FP8vLLM2K642,636.56 tok/s242.16
Qwen3-4B-FP8 ↗FP8vLLM2K1283,246.14 tok/s280.48
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K174.29 tok/s199.01
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K8572.12 tok/s215.36
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K16825.69 tok/s212.27
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K321,665.06 tok/s223.52
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K642,665.72 tok/s241.49
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K1283,194.65 tok/s279.21
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K2564,086.05 tok/s294.86
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K5124,301.45 tok/s296.96
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K174.1 tok/s197.73
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K8531.12 tok/s212.55
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K16933.97 tok/s208.24
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K321,645.04 tok/s218.77
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K642,653.11 tok/s239.51
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K1283,150.16 tok/s277.72
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K2564,074.31 tok/s293.27
Qwen3-4B-Instruct-FP8 ↗FP8vLLM2K5124,266.23 tok/s292.16
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K1246.18 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K2351.48 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K4578.86 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K8776.25 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K16778.28 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K1245.88 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K2351.95 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K4580.31 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K8778.01 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K16778.97 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K1246.24 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K2351.96 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K4578.91 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K8777.88 tok/s—
Qwen3-4B-Q4_K_M ↗Q4_K_MvLLM1K16780.32 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K1224.67 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K2319.79 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K4552.4 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K8765.63 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K16767.67 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K1225.21 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K2319.52 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K4550.33 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K8767.09 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K16769.14 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K1224.88 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K2320.07 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K4552.43 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K8767.56 tok/s—
Qwen3-4B-Q5_K_M ↗Q5_K_MvLLM1K16771.32 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K1170.23 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K2268.97 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K4486.39 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K8763.08 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K16766.16 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K1170.39 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K2269.12 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K4487.93 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K8762.92 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K16764.89 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K1170.27 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K2269.59 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K4488.85 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K8764.16 tok/s—
Qwen3-4B-Q8_0 ↗Q8_0vLLM1K16767.6 tok/s—
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K174.28 tok/s199.45
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K8569.24 tok/s214.92
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K16935.67 tok/s212.32
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K321,590.69 tok/s221.52
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K642,562.45 tok/s240.53
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K1283,075.68 tok/s280.2
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K2564,022.43 tok/s297.43
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K5124,200.93 tok/s298.09
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K173.36 tok/s193.18
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K8560.55 tok/s210.36
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K16918.65 tok/s206.99
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K321,607.47 tok/s218.44
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K642,566.37 tok/s238.09
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K1283,095.75 tok/s277.3
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K2564,073.56 tok/s297.38
Qwen3-4B-Thinking-FP8 ↗FP8vLLM2K5124,205.28 tok/s297.69
Qwen3-8B-AWQ ↗AWQvLLM2K1139.77 tok/s282.76
Qwen3-8B-AWQ ↗AWQvLLM2K8883.62 tok/s273.55
Qwen3-8B-AWQ ↗AWQvLLM2K161,273.62 tok/s269.7
Qwen3-8B-AWQ ↗AWQvLLM2K322,299.42 tok/s303.56
Qwen3-8B-AWQ ↗AWQvLLM2K642,999.88 tok/s316.37
Qwen3-8B-AWQ ↗AWQvLLM2K1283,016.02 tok/s316.05
Qwen3-8B-AWQ ↗AWQvLLM2K2563,432.32 tok/s314.78
Qwen3-8B-FP8 ↗FP8vLLM2K151.58 tok/s228.74
Qwen3-8B-FP8 ↗FP8vLLM2K8453.56 tok/s251.87
Qwen3-8B-FP8 ↗FP8vLLM2K16687.4 tok/s243.73
Qwen3-8B-FP8 ↗FP8vLLM2K321,361.86 tok/s251.1
Qwen3-8B-FP8 ↗FP8vLLM2K642,268.81 tok/s267.2
Qwen3-8B-FP8 ↗FP8vLLM2K1282,611.51 tok/s304.15
Qwen3-8B-FP8 ↗FP8vLLM2K2563,274.96 tok/s309.95
Qwen3-8B-FP8 ↗FP8vLLM2K151.42 tok/s229.09
Qwen3-8B-FP8 ↗FP8vLLM2K8458.38 tok/s253.52
Qwen3-8B-FP8 ↗FP8vLLM2K16687.22 tok/s247.51
Qwen3-8B-FP8 ↗FP8vLLM2K321,436.15 tok/s251.81
Qwen3-8B-FP8 ↗FP8vLLM2K642,351.02 tok/s269.89
Qwen3-8B-FP8 ↗FP8vLLM2K1282,599.07 tok/s305.71
Qwen3-8B-FP8 ↗FP8vLLM2K2563,371.27 tok/s311.44
Qwen3-8B-FP8 ↗FP8vLLM2K151.65 tok/s226.69
Qwen3-8B-FP8 ↗FP8vLLM2K8454.5 tok/s249.55
Qwen3-8B-FP8 ↗FP8vLLM2K16666.4 tok/s244.49
Qwen3-8B-FP8 ↗FP8vLLM2K321,426.13 tok/s250.87
Qwen3-8B-FP8 ↗FP8vLLM2K642,330.53 tok/s265.76
Qwen3-8B-FP8 ↗FP8vLLM2K1282,596.37 tok/s302.9
Qwen3-8B-FP8 ↗FP8vLLM2K2563,132.23 tok/s309.61
Qwen3-8B-FP8 ↗FP8vLLM2K151.51 tok/s219.25
Qwen3-8B-FP8 ↗FP8vLLM2K8450.16 tok/s252.08
Qwen3-8B-FP8 ↗FP8vLLM2K16760.97 tok/s245.93
Qwen3-8B-FP8 ↗FP8vLLM2K321,301.63 tok/s252.53
Qwen3-8B-FP8 ↗FP8vLLM2K642,369.35 tok/s267.96
Qwen3-8B-FP8 ↗FP8vLLM2K1282,554.99 tok/s305.69
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K1159.55 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K2252.98 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K4438.09 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K8573.83 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K16577.53 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K1159.16 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K2252.85 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K4438.26 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K8575.12 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K16578.08 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K1158.85 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K2250.76 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K4435.1 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K8569.16 tok/s—
Qwen3-8B-Q4_K_M ↗Q4_K_MvLLM1K16574.69 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K1141.66 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K2228.11 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K4411.62 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K8560.16 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K16564.72 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K1141.51 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K2228.82 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K4411.71 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K8562.98 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K16566.83 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K1141.38 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K2227.39 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K4406.49 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K8555.43 tok/s—
Qwen3-8B-Q5_K_M ↗Q5_K_MvLLM1K16560.68 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K1102.84 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K2175.67 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K4328.96 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K8550.2 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K16553.91 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K1102.85 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K2175.17 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K4329.71 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K8550.14 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K16554.33 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K1102.83 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K2175.36 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K4328.42 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K8543.25 tok/s—
Qwen3-8B-Q8_0 ↗Q8_0vLLM1K16549.67 tok/s—
Qwen3-14B-AWQ ↗AWQvLLM2K191.53 tok/s305.5
Qwen3-14B-AWQ ↗AWQvLLM2K8576.77 tok/s296.35
Qwen3-14B-AWQ ↗AWQvLLM2K16931.96 tok/s301.54
Qwen3-14B-AWQ ↗AWQvLLM2K321,539.11 tok/s339.88
Qwen3-14B-AWQ ↗AWQvLLM2K642,074.24 tok/s347.03
Qwen3-14B-AWQ ↗AWQvLLM2K1282,286.57 tok/s344.42
Qwen3-14B-AWQ ↗AWQvLLM2K2562,503.73 tok/s353.52
Qwen3-14B-FP8 ↗FP8vLLM2K132.8 tok/s247.48
Qwen3-14B-FP8 ↗FP8vLLM2K8229.88 tok/s267.57
Qwen3-14B-FP8 ↗FP8vLLM2K16407.24 tok/s263.18
Qwen3-14B-FP8 ↗FP8vLLM2K32726.89 tok/s267.84
Qwen3-14B-FP8 ↗FP8vLLM2K641,221.79 tok/s275.3
Qwen3-14B-FP8 ↗FP8vLLM2K1281,937.23 tok/s325.71
Qwen3-14B-FP8 ↗FP8vLLM2K2562,484.72 tok/s343.34
Qwen3-14B-FP8 ↗FP8vLLM2K132.87 tok/s246.53
Qwen3-14B-FP8 ↗FP8vLLM2K8248.37 tok/s264.54
Qwen3-14B-FP8 ↗FP8vLLM2K16418.77 tok/s260.19
Qwen3-14B-FP8 ↗FP8vLLM2K32718 tok/s264.95
Qwen3-14B-FP8 ↗FP8vLLM2K641,294.65 tok/s274.85
Qwen3-14B-FP8 ↗FP8vLLM2K1281,990.19 tok/s325.32
Qwen3-14B-FP8 ↗FP8vLLM2K2562,501.77 tok/s337.38
Qwen3-14B-FP8 ↗FP8vLLM2K132.94 tok/s244.41
Qwen3-14B-FP8 ↗FP8vLLM2K8223.92 tok/s261.52
Qwen3-14B-FP8 ↗FP8vLLM2K16393.72 tok/s259.58
Qwen3-14B-FP8 ↗FP8vLLM2K32748.12 tok/s261.87
Qwen3-14B-FP8 ↗FP8vLLM2K641,302.67 tok/s272.62
Qwen3-14B-FP8 ↗FP8vLLM2K1281,927.99 tok/s323.35
Qwen3-14B-FP8 ↗FP8vLLM2K2562,516.02 tok/s336.15
Qwen3-14B-FP8 ↗FP8vLLM2K132.77 tok/s238.03
Qwen3-14B-FP8 ↗FP8vLLM2K8239.71 tok/s265.89
Qwen3-14B-FP8 ↗FP8vLLM2K16373.01 tok/s262.2
Qwen3-14B-FP8 ↗FP8vLLM2K32708.79 tok/s266.13
Qwen3-14B-FP8 ↗FP8vLLM2K641,270.87 tok/s275.84
Qwen3-14B-FP8 ↗FP8vLLM2K1281,925.13 tok/s325.6
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K195.08 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K2162 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K4283.97 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K8370.9 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K16373.95 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K195.17 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K2162.04 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K4284.26 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K8370.11 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K16373.91 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K195.08 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K2161.17 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K4283.27 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K8370.35 tok/s—
Qwen3-14B-Q4_K_M ↗Q4_K_MvLLM1K16374.13 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K183.71 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K2144.19 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K4262.31 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K8362.14 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K16365.54 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K183.65 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K2144.16 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K4262 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K8361.12 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K16365.02 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K183.57 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K2143.79 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K4261.78 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K8361.66 tok/s—
Qwen3-14B-Q5_K_M ↗Q5_K_MvLLM1K16365.53 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K159.16 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K2106.56 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K4202.73 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K8351.33 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K16354.67 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K159.14 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K2106.73 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K4202.4 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K8350.9 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K16355.1 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K159.01 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K2106.01 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K4200.4 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K8348.1 tok/s—
Qwen3-14B-Q8_0 ↗Q8_0vLLM1K16352.72 tok/s—
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K1106.2 tok/s192.05
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K8406.92 tok/s245.31
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K16599.75 tok/s265.94
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K32968.6 tok/s265.2
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K641,596.36 tok/s273.4
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K1282,461.2 tok/s287.36
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K2563,357.9 tok/s290.56
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K5123,179.04 tok/s292.35
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K1106.2 tok/s191.4
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K8394.28 tok/s242.01
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K16634.03 tok/s259.33
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K32942.71 tok/s262.69
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K641,516.6 tok/s275.01
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K1282,472.09 tok/s285.22
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K2563,384.65 tok/s290.25
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K5123,197.39 tok/s289.54
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K1106.67 tok/s191.68
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K8394.57 tok/s247.15
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K16627.02 tok/s266.67
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K32994.12 tok/s268.01
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K641,362.38 tok/s273.38
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K1282,129.26 tok/s286.14
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K2562,334.15 tok/s296.53
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K1106.03 tok/s185.4
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K8399.3 tok/s249.16
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K16647.25 tok/s267.19
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K32975.24 tok/s268.89
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K641,352.57 tok/s277.38
Qwen3-30B-A3B-FP8 ↗FP8vLLM2K1282,156.26 tok/s288.11
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K1241.13 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K2186.16 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K4302.62 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K8469.7 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K16470.99 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K1240.92 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K2186.29 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K4303.71 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K8462.24 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K16466.04 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K1239.02 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K2183.24 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K4297.97 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K8452.95 tok/s—
Qwen3-30B-A3B-Q4_K_M ↗Q4_K_MvLLM1K16456.96 tok/s—
Qwen3-30B-A3B-Q8_0 ↗Q8_0vLLM1K1182.28 tok/s—
Qwen3-30B-A3B-Q8_0 ↗Q8_0vLLM1K2162.73 tok/s—
Qwen3-30B-A3B-Q8_0 ↗Q8_0vLLM1K4261.63 tok/s—
Qwen3-30B-A3B-Q8_0 ↗Q8_0vLLM1K8402.53 tok/s—
Qwen3-30B-A3B-Q8_0 ↗Q8_0vLLM1K16408.28 tok/s—
Qwen3-30B-A3B-Q8_0 ↗Q8_0vLLM1K1181.77 tok/s—
Qwen3-30B-A3B-Q8_0 ↗Q8_0vLLM1K2161.7 tok/s—
Qwen3-30B-A3B-Q8_0 ↗Q8_0vLLM1K4259.49 tok/s—
Qwen3-30B-A3B-Q8_0 ↗Q8_0vLLM1K8396.76 tok/s—
Qwen3-30B-A3B-Q8_0 ↗Q8_0vLLM1K16393.18 tok/s—
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K1106.38 tok/s193.41
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K8401.84 tok/s249.09
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K16607.23 tok/s266.07
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K32980.52 tok/s268.33
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K641,403.32 tok/s275.51
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K1282,108.81 tok/s289.3
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K2562,903.16 tok/s294.06
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K1106.34 tok/s191.64
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K8399.46 tok/s245.39
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K16623.26 tok/s265.26
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K32996.01 tok/s267.35
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K641,372.82 tok/s276.92
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K1282,107.95 tok/s290.21
Qwen3-30B-A3B-Thinking-FP8 ↗FP8vLLM2K2563,003.25 tok/s292.33
Qwen3-32B-AWQ ↗AWQvLLM2K146.34 tok/s323.63
Qwen3-32B-AWQ ↗AWQvLLM2K8292.18 tok/s315.33
Qwen3-32B-AWQ ↗AWQvLLM2K16475.16 tok/s332.09
Qwen3-32B-AWQ ↗AWQvLLM2K32706.34 tok/s365.23
Qwen3-32B-AWQ ↗AWQvLLM2K641,028 tok/s369.03
Qwen3-32B-AWQ ↗AWQvLLM2K1281,160.53 tok/s369.83
Qwen3-32B-FP8 ↗FP8vLLM2K116.17 tok/s262.49
Qwen3-32B-FP8 ↗FP8vLLM2K8116.06 tok/s286.72
Qwen3-32B-FP8 ↗FP8vLLM2K16183.12 tok/s284.63
Qwen3-32B-FP8 ↗FP8vLLM2K32406.64 tok/s282.73
Qwen3-32B-FP8 ↗FP8vLLM2K64687.14 tok/s289.26
Qwen3-32B-FP8 ↗FP8vLLM2K116.2 tok/s259.91
Qwen3-32B-FP8 ↗FP8vLLM2K8136.83 tok/s283.86
Qwen3-32B-FP8 ↗FP8vLLM2K16182.58 tok/s277.89
Qwen3-32B-FP8 ↗FP8vLLM2K32371.07 tok/s284.36
Qwen3-32B-FP8 ↗FP8vLLM2K64676.99 tok/s287.63
Qwen3-32B-FP8 ↗FP8vLLM2K116.2 tok/s258.68
Qwen3-32B-FP8 ↗FP8vLLM2K8136.95 tok/s280.71
Qwen3-32B-FP8 ↗FP8vLLM2K16181.55 tok/s279.15
Qwen3-32B-FP8 ↗FP8vLLM2K32372.39 tok/s280.54
Qwen3-32B-FP8 ↗FP8vLLM2K64443.99 tok/s286.02
Qwen3-32B-FP8 ↗FP8vLLM2K116.09 tok/s251
Qwen3-32B-FP8 ↗FP8vLLM2K8113.89 tok/s284.17
Qwen3-32B-FP8 ↗FP8vLLM2K16186.28 tok/s281.54
Qwen3-32B-FP8 ↗FP8vLLM2K32378.01 tok/s284.42
Qwen3-32B-FP8 ↗FP8vLLM2K64468.56 tok/s290.4
Qwen3-32B-Q4_K_M ↗Q4_K_MvLLM1K144.09 tok/s—
Qwen3-32B-Q4_K_M ↗Q4_K_MvLLM1K277.87 tok/s—
Qwen3-32B-Q4_K_M ↗Q4_K_MvLLM1K4138.01 tok/s—
Qwen3-32B-Q4_K_M ↗Q4_K_MvLLM1K8176.07 tok/s—
Qwen3-32B-Q4_K_M ↗Q4_K_MvLLM1K16179.94 tok/s—
Qwen3-32B-Q4_K_M ↗Q4_K_MvLLM1K144.09 tok/s—
Qwen3-32B-Q4_K_M ↗Q4_K_MvLLM1K277.92 tok/s—
Qwen3-32B-Q4_K_M ↗Q4_K_MvLLM1K4137.74 tok/s—
Qwen3-32B-Q4_K_M ↗Q4_K_MvLLM1K8175.89 tok/s—
Qwen3-32B-Q4_K_M ↗Q4_K_MvLLM1K16180.01 tok/s—
Qwen3-32B-Q5_K_M ↗Q5_K_MvLLM1K138.27 tok/s—
Qwen3-32B-Q5_K_M ↗Q5_K_MvLLM1K268.31 tok/s—
Qwen3-32B-Q5_K_M ↗Q5_K_MvLLM1K4125.49 tok/s—
Qwen3-32B-Q5_K_M ↗Q5_K_MvLLM1K8170.59 tok/s—
Qwen3-32B-Q5_K_M ↗Q5_K_MvLLM1K16174.51 tok/s—
Qwen3-VL-2B-FP8 ↗FP8vLLM2K1135.49 tok/s180.04
Qwen3-VL-2B-FP8 ↗FP8vLLM2K8967.34 tok/s196.55
Qwen3-VL-2B-FP8 ↗FP8vLLM2K161,612.34 tok/s190.12
Qwen3-VL-2B-FP8 ↗FP8vLLM2K322,623.08 tok/s207.92
Qwen3-VL-2B-FP8 ↗FP8vLLM2K643,769.38 tok/s231.58
Qwen3-VL-2B-FP8 ↗FP8vLLM2K1285,031.73 tok/s263.84
Qwen3-VL-2B-FP8 ↗FP8vLLM2K2565,471.64 tok/s277.59
Qwen3-VL-2B-FP8 ↗FP8vLLM2K5125,738.93 tok/s278.59
Qwen3-VL-2B-FP8 ↗FP8vLLM2K1135.58 tok/s179.28
Qwen3-VL-2B-FP8 ↗FP8vLLM2K8975.63 tok/s194.74
Qwen3-VL-2B-FP8 ↗FP8vLLM2K161,562.54 tok/s189.88
Qwen3-VL-2B-FP8 ↗FP8vLLM2K322,628.95 tok/s206.37
Qwen3-VL-2B-FP8 ↗FP8vLLM2K643,789.86 tok/s233.5
Qwen3-VL-2B-FP8 ↗FP8vLLM2K1285,044.73 tok/s264.21
Qwen3-VL-2B-FP8 ↗FP8vLLM2K2565,519.06 tok/s279.59
Qwen3-VL-2B-FP8 ↗FP8vLLM2K5125,709.94 tok/s277.06
Qwen3-VL-2B-FP8 ↗FP8vLLM2K1134.43 tok/s176.42
Qwen3-VL-2B-FP8 ↗FP8vLLM2K8926.86 tok/s193.14
Qwen3-VL-2B-FP8 ↗FP8vLLM2K161,575.67 tok/s186.03
Qwen3-VL-2B-FP8 ↗FP8vLLM2K322,590.63 tok/s204.2
Qwen3-VL-2B-FP8 ↗FP8vLLM2K643,715.75 tok/s231.34
Qwen3-VL-2B-FP8 ↗FP8vLLM2K1285,063.88 tok/s263.3
Qwen3-VL-2B-FP8 ↗FP8vLLM2K2565,523.21 tok/s277.02
Qwen3-VL-2B-FP8 ↗FP8vLLM2K5125,754.84 tok/s276.92
Qwen3-VL-2B-FP8 ↗FP8vLLM2K1136.31 tok/s178.28
Qwen3-VL-2B-FP8 ↗FP8vLLM2K8954.81 tok/s193.76
Qwen3-VL-2B-FP8 ↗FP8vLLM2K161,611.31 tok/s188.05
Qwen3-VL-2B-FP8 ↗FP8vLLM2K322,636.18 tok/s205.06
Qwen3-VL-2B-FP8 ↗FP8vLLM2K643,720.63 tok/s230.27
Qwen3-VL-2B-FP8 ↗FP8vLLM2K1285,022.91 tok/s261.06
Qwen3-VL-2B-FP8 ↗FP8vLLM2K2565,534.9 tok/s276.54
Qwen3-VL-4B-FP8 ↗FP8vLLM2K173.66 tok/s198.29
Qwen3-VL-4B-FP8 ↗FP8vLLM2K8567.5 tok/s210.54
Qwen3-VL-4B-FP8 ↗FP8vLLM2K161,006.86 tok/s207.2
Qwen3-VL-4B-FP8 ↗FP8vLLM2K321,678.88 tok/s216.37
Qwen3-VL-4B-FP8 ↗FP8vLLM2K642,554.3 tok/s233.73
Qwen3-VL-4B-FP8 ↗FP8vLLM2K1283,404.39 tok/s273.65
Qwen3-VL-4B-FP8 ↗FP8vLLM2K2564,185.57 tok/s293.15
Qwen3-VL-4B-FP8 ↗FP8vLLM2K5124,231.81 tok/s292.9
Qwen3-VL-4B-FP8 ↗FP8vLLM2K173.83 tok/s196.81
Qwen3-VL-4B-FP8 ↗FP8vLLM2K8541.59 tok/s210.75
Qwen3-VL-4B-FP8 ↗FP8vLLM2K16968.67 tok/s205.21
Qwen3-VL-4B-FP8 ↗FP8vLLM2K321,519.97 tok/s214.53
Qwen3-VL-4B-FP8 ↗FP8vLLM2K642,477.56 tok/s233.58
Qwen3-VL-4B-FP8 ↗FP8vLLM2K1283,436.23 tok/s272.94
Qwen3-VL-4B-FP8 ↗FP8vLLM2K2564,139.24 tok/s292.68
Qwen3-VL-4B-FP8 ↗FP8vLLM2K5124,199.45 tok/s291.88
Qwen3-VL-4B-FP8 ↗FP8vLLM2K173.01 tok/s193.84
Qwen3-VL-4B-FP8 ↗FP8vLLM2K8498.46 tok/s207.7
Qwen3-VL-4B-FP8 ↗FP8vLLM2K16874.73 tok/s201.65
Qwen3-VL-4B-FP8 ↗FP8vLLM2K321,638.15 tok/s212.69
Qwen3-VL-4B-FP8 ↗FP8vLLM2K642,534.64 tok/s232.52
Qwen3-VL-4B-FP8 ↗FP8vLLM2K1283,419.49 tok/s270.99
Qwen3-VL-4B-FP8 ↗FP8vLLM2K2564,126.67 tok/s292.11
Qwen3-VL-8B-FP8 ↗FP8vLLM2K151 tok/s227.48
Qwen3-VL-8B-FP8 ↗FP8vLLM2K8387.07 tok/s248.16
Qwen3-VL-8B-FP8 ↗FP8vLLM2K16706.33 tok/s241.71
Qwen3-VL-8B-FP8 ↗FP8vLLM2K321,229.75 tok/s251.18
Qwen3-VL-8B-FP8 ↗FP8vLLM2K641,975.17 tok/s262.92
Qwen3-VL-8B-FP8 ↗FP8vLLM2K1282,768.2 tok/s302.62
Qwen3-VL-8B-FP8 ↗FP8vLLM2K2563,406.27 tok/s311
Qwen3-VL-8B-FP8 ↗FP8vLLM2K5123,479.51 tok/s309.82
Qwen3-VL-8B-FP8 ↗FP8vLLM2K151.27 tok/s225.16
Qwen3-VL-8B-FP8 ↗FP8vLLM2K8356.8 tok/s246.18
Qwen3-VL-8B-FP8 ↗FP8vLLM2K16640.69 tok/s240.22
Qwen3-VL-8B-FP8 ↗FP8vLLM2K321,162.31 tok/s248.22
Qwen3-VL-8B-FP8 ↗FP8vLLM2K641,963.24 tok/s262.14
Qwen3-VL-8B-FP8 ↗FP8vLLM2K1282,800.56 tok/s301.17
Qwen3-VL-8B-FP8 ↗FP8vLLM2K2563,411.53 tok/s307.65
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K1103.36 tok/s184.14
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K8355.42 tok/s241.84
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K16550.36 tok/s259.99
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K32860.65 tok/s261.52
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K641,389.92 tok/s273.04
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K1282,166.93 tok/s285.3
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K2562,879.43 tok/s288.82
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K5122,901.88 tok/s286.26
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K1103.88 tok/s188.49
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K8358.75 tok/s249.5
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K16555.94 tok/s263.88
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K32882 tok/s262.59
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K641,342.16 tok/s275.16
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K1105.51 tok/s189.72
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K8355.07 tok/s245.27
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K16583.48 tok/s257.81
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K32838.91 tok/s261.45
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K641,355.61 tok/s274.82
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K1282,198.82 tok/s284.83
Qwen3-VL-30B-A3B-FP8 ↗FP8vLLM2K2562,851.94 tok/s285.61
Qwen3-VL-32B-FP8 ↗FP8vLLM2K116.27 tok/s258.29
Qwen3-VL-32B-FP8 ↗FP8vLLM2K8121.1 tok/s280.53
Qwen3-VL-32B-FP8 ↗FP8vLLM2K16198.1 tok/s278.39
Qwen3-VL-32B-FP8 ↗FP8vLLM2K32438.17 tok/s278.61
Qwen3-VL-32B-FP8 ↗FP8vLLM2K64689.51 tok/s292.41
Qwen3-VL-32B-FP8 ↗FP8vLLM2K128847.58 tok/s331.13
Qwen3-VL-32B-FP8 ↗FP8vLLM2K256889.78 tok/s353.01
Qwen3-VL-32B-FP8 ↗FP8vLLM2K5121,254.31 tok/s349.3
Qwen3-VL-32B-FP8 ↗FP8vLLM2K116.24 tok/s261.07
Qwen3-VL-32B-FP8 ↗FP8vLLM2K8110.42 tok/s283.05
Qwen3-VL-32B-FP8 ↗FP8vLLM2K116.21 tok/s258.88
Qwen3-VL-32B-FP8 ↗FP8vLLM2K8133.11 tok/s279.95
Qwen3-VL-32B-FP8 ↗FP8vLLM2K16213.03 tok/s277.63
Qwen3-VL-32B-FP8 ↗FP8vLLM2K32459.21 tok/s277.09
Qwen3-VL-32B-FP8 ↗FP8vLLM2K64571.37 tok/s291.69
Qwen3-VL-32B-FP8 ↗FP8vLLM2K128690.52 tok/s331.63
Qwen3-VL-32B-FP8 ↗FP8vLLM2K256801.87 tok/s346.85

Original reports

Every report for this task, including quality, diagnostic and legacy runs that do not enter comparisons.
Original reports · 445 rows
ModelTest taskFormatEngineWorkloadProtocolQuality scoreTest date
Gemma-3-270m ↗Text generationBF16vLLM1KMethod unknown—3/6/2026, 10:08:36 PM
Gemma-3-270m ↗Text generationBF16vLLM1KMethod unknown—3/6/2026, 8:39:50 PM
Gemma-3-270m ↗Text generationBF16vLLM1KMethod unknown—3/6/2026, 8:09:47 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM1KMethod unknown—2/7/2026, 2:34:31 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM1KMethod unknown—1/25/2026, 5:14:18 PM
Qwen3-8B-AWQ ↗Text generationAWQvLLM1KMethod unknown—1/25/2026, 7:10:33 AM
Qwen3-4B-AWQ ↗Text generationAWQvLLM1KMethod unknown—1/25/2026, 6:32:49 AM
Gemma-3-270m ↗Text generationBF16vLLM1KMethod unknown—1/25/2026, 6:03:52 AM
gemma-3-270m ↗Text generationNativevLLM1KMethod unknown—1/25/2026, 4:33:01 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM1KMethod unknown—1/25/2026, 3:22:58 AM
Qwen3-1.7B ↗Text generationBF16vLLM1KMethod unknown—1/22/2026, 9:52:43 PM
Qwen3-4B ↗Text generationBF16vLLM1KMethod unknown—1/22/2026, 9:44:00 PM
Qwen3-4B-AWQ ↗Text generationAWQvLLM1KMethod unknown—1/22/2026, 8:27:13 PM
HY-MT1.5-1.8B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/22/2026, 4:41:06 AM
HY-MT1.5-1.8B-Q6_K ↗Text generationQ6_KvLLM1KMethod unknown—1/22/2026, 4:39:13 AM
HY-MT1.5-1.8B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/22/2026, 4:37:21 AM
Qwen3-1.7B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/22/2026, 4:35:30 AM
Qwen3-0.6B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/22/2026, 4:33:38 AM
Qwen3-0.6B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/21/2026, 12:05:51 AM
Qwen3-30B-A3B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/21/2026, 12:02:04 AM
Qwen3-14B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/21/2026, 12:00:07 AM
Qwen3-14B-Q5_K_M ↗Text generationQ5_K_MvLLM1KMethod unknown—1/20/2026, 11:58:08 PM
Qwen3-14B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 11:56:08 PM
Qwen3-8B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 11:54:10 PM
Qwen3-8B-Q5_K_M ↗Text generationQ5_K_MvLLM1KMethod unknown—1/20/2026, 11:52:14 PM
Qwen3-8B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 11:50:20 PM
HY-MT1.5-7B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 11:48:28 PM
HY-MT1.5-7B-Q6_K ↗Text generationQ6_KvLLM1KMethod unknown—1/20/2026, 11:46:34 PM
HY-MT1.5-7B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 11:44:39 PM
Qwen3-4B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 11:42:47 PM
Qwen3-4B-Q5_K_M ↗Text generationQ5_K_MvLLM1KMethod unknown—1/20/2026, 11:40:55 PM
Qwen3-4B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 11:39:04 PM
HY-MT1.5-1.8B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 11:37:10 PM
HY-MT1.5-1.8B-Q6_K ↗Text generationQ6_KvLLM1KMethod unknown—1/20/2026, 11:35:20 PM
HY-MT1.5-1.8B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 11:33:27 PM
Qwen3-1.7B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 11:31:38 PM
Qwen3-0.6B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 11:29:48 PM
Qwen3-32B-Q5_K_M ↗Text generationQ5_K_MvLLM1KMethod unknown—1/20/2026, 8:02:21 PM
Qwen3-32B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 8:00:15 PM
Qwen3-30B-A3B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 7:58:10 PM
Qwen3-30B-A3B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 7:56:10 PM
Qwen3-14B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 7:54:11 PM
Qwen3-14B-Q5_K_M ↗Text generationQ5_K_MvLLM1KMethod unknown—1/20/2026, 7:52:11 PM
Qwen3-14B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 7:50:11 PM
Qwen3-8B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 7:48:15 PM
Qwen3-8B-Q5_K_M ↗Text generationQ5_K_MvLLM1KMethod unknown—1/20/2026, 7:46:21 PM
Qwen3-8B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 7:44:28 PM
HY-MT1.5-7B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 7:42:35 PM
HY-MT1.5-7B-Q6_K ↗Text generationQ6_KvLLM1KMethod unknown—1/20/2026, 7:40:39 PM
HY-MT1.5-7B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 7:38:44 PM
Qwen3-4B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 7:36:44 PM
Qwen3-4B-Q5_K_M ↗Text generationQ5_K_MvLLM1KMethod unknown—1/20/2026, 7:34:50 PM
Qwen3-4B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 7:32:56 PM
HY-MT1.5-1.8B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 7:31:06 PM
HY-MT1.5-1.8B-Q6_K ↗Text generationQ6_KvLLM1KMethod unknown—1/20/2026, 7:29:13 PM
HY-MT1.5-1.8B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 7:27:24 PM
Qwen3-1.7B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 7:25:35 PM
Qwen3-0.6B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 7:23:44 PM
Qwen3-32B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 5:05:10 PM
Qwen3-30B-A3B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 4:56:37 PM
Qwen3-30B-A3B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 4:22:46 PM
Qwen3-14B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 4:13:41 PM
Qwen3-14B-Q5_K_M ↗Text generationQ5_K_MvLLM1KMethod unknown—1/20/2026, 4:07:27 PM
Qwen3-14B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 4:02:01 PM
Qwen3-8B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 3:57:45 PM
Qwen3-8B-Q5_K_M ↗Text generationQ5_K_MvLLM1KMethod unknown—1/20/2026, 3:54:34 PM
Qwen3-8B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 3:52:22 PM
HY-MT1.5-7B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 3:50:25 PM
HY-MT1.5-7B-Q6_K ↗Text generationQ6_KvLLM1KMethod unknown—1/20/2026, 3:46:40 PM
HY-MT1.5-7B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 3:44:03 PM
Qwen3-4B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 3:42:08 PM
Qwen3-4B-Q5_K_M ↗Text generationQ5_K_MvLLM1KMethod unknown—1/20/2026, 3:40:15 PM
Qwen3-4B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 3:38:19 PM
HY-MT1.5-1.8B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 3:36:24 PM
HY-MT1.5-1.8B-Q6_K ↗Text generationQ6_KvLLM1KMethod unknown—1/20/2026, 3:34:32 PM
HY-MT1.5-1.8B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 3:32:41 PM
Qwen3-1.7B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 3:30:49 PM
Qwen3-0.6B-Q8_0 ↗Text generationQ8_0vLLM1KMethod unknown—1/20/2026, 3:28:52 PM
Qwen2.5-0.5B-Q4_K_M ↗Text generationQ4_K_MvLLM1KMethod unknown—1/20/2026, 11:24:33 AM
Gemma-3-270m ↗Text generationBF16vLLM2KMethod unknown—3/6/2026, 10:15:08 PM
Gemma-3-270m ↗Text generationBF16vLLM2KMethod unknown—3/6/2026, 8:46:56 PM
Gemma-3-270m ↗Text generationBF16vLLM2KMethod unknown—3/6/2026, 8:17:37 PM
DeepSeek-R1-Distill-Llama-8B ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 6:28:03 PM
DeepSeek-R1-Distill-Qwen-7B ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 5:56:49 PM
DASD-4B-Thinking ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 5:25:35 PM
Nanbeige4-3B-Thinking-2511 ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 4:56:30 PM
DeepSeek-R1-Distill-Qwen-1.5B ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 4:37:38 PM
AutoGLM-Phone-9B ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 11:34:43 AM
Step3-VL-10B ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 10:44:11 AM
MiniCPM-o-4_5 ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 9:54:41 AM
MAI-UI-8B ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 9:06:04 AM
dots.ocr ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 7:54:51 AM
DeepSeek-OCR-2 ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 7:31:44 AM
HunyuanOCR ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 6:57:44 AM
LightOnOCR-2-1B ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 6:35:52 AM
PaddleOCR-VL-1.5 ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 6:12:32 AM
PaddleOCR-VL-1.5 ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 6:06:59 AM
DeepSeek-R1-Distill-Qwen-1.5B ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 5:42:06 AM
LightOnOCR-2-1B ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 5:30:22 AM
PaddleOCR-VL-1.5 ↗Text generationNativevLLM2KMethod unknown—3/4/2026, 4:34:39 AM
Qwen3.5-9B ↗Text generationNativevLLM2KMethod unknown—3/3/2026, 9:39:06 AM
Qwen3.5-4B ↗Text generationNativevLLM2KMethod unknown—3/3/2026, 9:32:08 AM
Qwen3.5-2B ↗Text generationNativevLLM2KMethod unknown—3/3/2026, 9:23:57 AM
Qwen3.5-0.8B ↗Text generationNativevLLM2KMethod unknown—3/3/2026, 9:16:20 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—3/2/2026, 10:19:54 PM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—3/2/2026, 9:19:18 PM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—3/2/2026, 9:05:44 PM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—3/1/2026, 3:52:51 PM
Qwen3.5-27B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/27/2026, 12:41:19 PM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/11/2026, 2:14:25 AM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 6:31:18 PM
Qwen3-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 6:12:14 PM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 6:03:20 PM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 5:53:56 PM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 5:44:50 PM
Qwen3-14B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 5:35:46 PM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 5:26:36 PM
Qwen3-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 5:18:05 PM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 5:10:28 PM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 5:02:40 PM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 4:55:10 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 4:47:03 PM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/9/2026, 4:38:34 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 9:26:36 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 9:19:15 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 9:12:18 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 9:05:17 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 8:36:17 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 7:59:43 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 7:52:31 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 7:46:23 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 7:40:38 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 7:33:24 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 7:09:50 AM
Qwen2.5-1.5B-Instruct-AWQ ↗Text generationAWQvLLM2KMethod unknown—2/7/2026, 6:46:46 AM
Qwen2.5-1.5B-Instruct-AWQ ↗Text generationAWQvLLM2KMethod unknown—2/7/2026, 6:36:02 AM
Qwen2.5-1.5B-Instruct-AWQ ↗Text generationAWQvLLM2KMethod unknown—2/7/2026, 6:24:22 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 2:40:19 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/7/2026, 1:29:44 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/6/2026, 6:41:49 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/6/2026, 6:37:22 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/6/2026, 6:30:47 AM
Qwen2.5-7B-Instruct-AWQ ↗Text generationAWQvLLM2KMethod unknown—2/6/2026, 5:25:17 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/3/2026, 1:35:40 PM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—2/3/2026, 1:19:22 PM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 5:07:41 PM
Qwen3-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 3:48:31 PM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 3:03:08 PM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 2:18:48 PM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 1:01:39 PM
Qwen3-14B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 11:32:13 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 10:45:38 AM
Qwen3-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 9:35:21 AM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 8:42:16 AM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 7:44:54 AM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 6:54:55 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 5:51:00 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 5:24:50 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 5:02:00 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 4:40:56 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 4:09:27 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 3:50:36 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 3:25:08 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 3:07:30 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/26/2026, 3:06:33 AM
Qwen3-8B-AWQ ↗Text generationAWQvLLM2KMethod unknown—1/25/2026, 7:20:41 AM
gemma-3-270m ↗Text generationNativevLLM2KMethod unknown—1/25/2026, 4:47:46 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/25/2026, 3:33:08 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/23/2026, 12:24:49 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/23/2026, 12:07:35 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 11:47:23 PM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 11:31:08 PM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 11:18:25 AM
Qwen3-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 10:34:40 AM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 10:16:20 AM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 10:01:13 AM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 9:45:50 AM
Qwen3-14B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 9:28:52 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 9:15:47 AM
Qwen3-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 9:00:30 AM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 8:47:46 AM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 8:34:36 AM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 8:23:18 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 8:09:57 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 7:49:56 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 7:26:47 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 7:06:39 AM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 6:50:34 AM
Qwen3-14B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 6:33:27 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 6:20:23 AM
Qwen3-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 6:05:20 AM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 5:52:24 AM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 5:39:27 AM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 5:28:07 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 5:14:57 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/22/2026, 4:54:52 AM
Qwen3-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 6:01:33 AM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 5:43:44 AM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 5:28:14 AM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 5:12:46 AM
Qwen3-14B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 4:55:39 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 4:42:24 AM
Qwen3-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 4:27:17 AM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 4:14:27 AM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 4:01:14 AM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 3:49:49 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 3:36:34 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 3:17:00 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 2:14:41 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 1:54:48 AM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 1:33:14 AM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 1:18:25 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 1:04:48 AM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/20/2026, 12:53:18 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 4:26:39 PM
Qwen3-0.6B ↗Text generationBF16vLLM2KMethod unknown—1/15/2026, 3:37:15 PM
Qwen3-1.7B ↗Text generationBF16vLLM2KMethod unknown—1/15/2026, 3:23:17 PM
Qwen3-4B ↗Text generationBF16vLLM2KMethod unknown—1/15/2026, 3:11:58 PM
Qwen3-32B-AWQ ↗Text generationAWQvLLM2KMethod unknown—1/15/2026, 2:56:53 PM
Qwen3-14B-AWQ ↗Text generationAWQvLLM2KMethod unknown—1/15/2026, 2:44:05 PM
Qwen3-8B-AWQ ↗Text generationAWQvLLM2KMethod unknown—1/15/2026, 2:32:20 PM
Qwen3-4B-AWQ ↗Text generationAWQvLLM2KMethod unknown—1/15/2026, 2:21:39 PM
Qwen3-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 11:46:15 AM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 11:33:06 AM
Qwen3-14B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 11:19:25 AM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 10:55:10 AM
Qwen3-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 10:50:17 AM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 10:38:39 AM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 10:26:57 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 10:15:35 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 10:01:35 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/15/2026, 9:46:46 AM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/14/2026, 9:16:49 PM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/14/2026, 9:04:30 PM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/14/2026, 8:52:10 PM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/14/2026, 8:40:29 PM
Qwen2.5-72B-AWQ ↗Text generationAWQvLLM2KMethod unknown—1/14/2026, 8:15:40 AM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/14/2026, 4:56:43 AM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/14/2026, 4:34:55 AM
Qwen3-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/14/2026, 4:14:36 AM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/14/2026, 3:16:16 AM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 11:57:27 AM
Qwen3-14B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 11:51:54 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 11:38:12 AM
Qwen3-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 11:26:08 AM
Hunyuan-7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 11:14:03 AM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 11:02:52 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 10:52:27 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 10:40:59 AM
Hunyuan-1.8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 10:28:46 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 10:17:22 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 10:05:45 AM
Hunyuan-7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 1:34:54 AM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 1:30:03 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 1:21:54 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 1:12:54 AM
Hunyuan-1.8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 1:05:02 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 12:58:00 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/13/2026, 12:50:21 AM
Qwen2.5-72B-AWQ ↗Text generationAWQvLLM2KMethod unknown—1/12/2026, 8:12:35 PM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 7:53:41 PM
Qwen3-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 7:44:37 PM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 7:30:30 PM
Qwen3-14B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 7:22:25 PM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 7:13:50 PM
Qwen3-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 7:06:09 PM
Hunyuan-7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 6:58:29 PM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 4:14:27 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 4:07:23 PM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 4:00:25 PM
Hunyuan-1.8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 3:53:51 PM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 3:47:52 PM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 3:41:11 PM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/12/2026, 1:35:23 PM
Qwen3-32B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/10/2026, 5:40:41 PM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/10/2026, 5:29:56 PM
Qwen3-14B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/10/2026, 5:23:48 PM
Qwen3-8B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/10/2026, 5:17:29 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/10/2026, 5:12:11 PM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/10/2026, 5:07:24 PM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM2KMethod unknown—1/10/2026, 5:03:04 PM
Gemma-3-270m ↗Text generationBF16vLLM4KMethod unknown—3/6/2026, 10:21:36 PM
Gemma-3-270m ↗Text generationBF16vLLM4KMethod unknown—3/6/2026, 8:55:28 PM
Gemma-3-270m ↗Text generationBF16vLLM4KMethod unknown—3/6/2026, 8:26:23 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM4KMethod unknown—2/7/2026, 2:46:49 AM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 5:20:55 PM
Qwen3-32B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 4:10:15 PM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 3:15:33 PM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 2:31:20 PM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 1:13:43 PM
Qwen3-14B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 11:43:46 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 10:55:04 AM
Qwen3-8B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 9:45:07 AM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 8:50:14 AM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 7:53:24 AM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 7:03:19 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 5:59:22 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 5:34:10 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 5:11:05 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 4:50:09 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 4:17:08 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 3:57:42 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 3:32:08 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/26/2026, 3:14:02 AM
Qwen3-8B-AWQ ↗Text generationAWQvLLM4KMethod unknown—1/25/2026, 7:35:48 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM4KMethod unknown—1/25/2026, 3:49:49 AM
Qwen3-32B-AWQ ↗Text generationAWQvLLM4KMethod unknown—1/22/2026, 8:48:34 PM
Gemma-3-270m ↗Text generationBF16vLLM8KMethod unknown—3/6/2026, 10:29:22 PM
Gemma-3-270m ↗Text generationBF16vLLM8KMethod unknown—3/6/2026, 9:01:54 PM
Gemma-3-270m ↗Text generationBF16vLLM8KMethod unknown—3/6/2026, 8:34:40 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM8KMethod unknown—2/7/2026, 2:59:36 AM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 5:43:05 PM
Qwen3-32B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 4:50:40 PM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 3:31:53 PM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 2:47:10 PM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 1:26:59 PM
Qwen3-14B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 11:58:25 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 11:06:07 AM
Qwen3-8B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 10:01:31 AM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 9:03:58 AM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 8:07:12 AM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 7:16:41 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 6:13:24 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 5:49:07 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 5:23:39 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 5:00:40 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 4:29:14 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 4:08:08 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 3:41:45 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/26/2026, 3:23:28 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM8KMethod unknown—1/25/2026, 1:49:45 AM
Gemma-3-270m ↗Text generationBF16vLLM16KMethod unknown—3/6/2026, 10:39:20 PM
Gemma-3-270m ↗Text generationBF16vLLM16KMethod unknown—3/6/2026, 8:40:32 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM16KMethod unknown—2/7/2026, 3:34:14 AM
Gemma-3-270m ↗Text generationBF16vLLM32KMethod unknown—3/6/2026, 10:43:08 PM
Gemma-3-270m ↗Text generationBF16vLLM32KMethod unknown—3/6/2026, 8:45:10 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM32KMethod unknown—2/7/2026, 3:50:21 AM
Gemma-3-270m ↗Text generationBF16vLLM64KMethod unknown—3/6/2026, 10:47:54 PM
Gemma-3-270m ↗Text generationBF16vLLM64KMethod unknown—3/6/2026, 8:48:24 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM64KMethod unknown—2/7/2026, 4:07:59 AM
Gemma-3-270m ↗Text generationBF16vLLM128Method unknown—3/6/2026, 9:25:21 PM
Gemma-3-270m ↗Text generationBF16vLLM128Method unknown—3/6/2026, 9:06:52 PM
Gemma-3-270m ↗Text generationBF16vLLM128Method unknown—3/6/2026, 7:53:02 PM
Gemma-3-270m ↗Text generationBF16vLLM128Method unknown—3/6/2026, 7:43:39 PM
Gemma-3-270m ↗Text generationBF16vLLM128Method unknown—3/6/2026, 4:29:01 PM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM128Method unknown—2/10/2026, 10:28:01 AM
Qwen2.5-1.5B-Instruct-AWQ ↗Text generationAWQvLLM128Method unknown—2/7/2026, 6:52:03 AM
Qwen2.5-0.5B-Instruct-AWQ ↗Text generationAWQvLLM128Method unknown—2/7/2026, 4:24:16 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM128Method unknown—2/7/2026, 2:14:44 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM128Method unknown—1/26/2026, 1:29:36 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM128Method unknown—1/25/2026, 4:37:27 PM
Qwen3-8B-AWQ ↗Text generationAWQvLLM128Method unknown—1/25/2026, 6:45:32 AM
Qwen3-4B-AWQ ↗Text generationAWQvLLM128Method unknown—1/25/2026, 6:10:53 AM
Gemma-3-270m ↗Text generationBF16vLLM128Method unknown—1/25/2026, 5:07:39 AM
gemma-3-270m ↗Text generationNativevLLM128Method unknown—1/25/2026, 4:07:02 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM128Method unknown—1/25/2026, 3:00:26 AM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM128Method unknown—1/23/2026, 3:53:56 AM
Qwen3-0.6B ↗Text generationBF16vLLM128Method unknown—1/23/2026, 3:39:14 AM
Qwen3-1.7B ↗Text generationBF16vLLM128Method unknown—1/23/2026, 3:31:47 AM
Qwen3-4B ↗Text generationBF16vLLM128Method unknown—1/23/2026, 3:25:44 AM
Qwen3-8B-AWQ ↗Text generationAWQvLLM128Method unknown—1/23/2026, 3:03:39 AM
Qwen3-4B-AWQ ↗Text generationAWQvLLM128Method unknown—1/23/2026, 2:57:30 AM
Qwen3-14B-AWQ ↗Text generationAWQvLLM128Method unknown—1/22/2026, 8:42:05 PM
Gemma-3-270m ↗Text generationBF16vLLM128KMethod unknown—3/6/2026, 10:51:17 PM
Gemma-3-270m ↗Text generationBF16vLLM128KMethod unknown—3/6/2026, 8:51:38 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM128KMethod unknown—2/7/2026, 4:11:10 AM
Gemma-3-270m ↗Text generationBF16vLLM256Method unknown—3/6/2026, 9:50:42 PM
Gemma-3-270m ↗Text generationBF16vLLM256Method unknown—3/6/2026, 8:16:37 PM
Gemma-3-270m ↗Text generationBF16vLLM256Method unknown—3/6/2026, 7:52:24 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM256Method unknown—2/7/2026, 2:22:07 AM
Qwen3-32B-AWQ ↗Text generationAWQvLLM256Method unknown—2/6/2026, 1:49:37 PM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 1:44:06 PM
Qwen3-32B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 1:38:23 PM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 1:32:34 PM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 1:27:00 PM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 1:21:01 PM
Qwen3-14B-AWQ ↗Text generationAWQvLLM256Method unknown—2/6/2026, 1:14:50 PM
Qwen3-14B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 1:08:55 PM
Qwen3-8B-AWQ ↗Text generationAWQvLLM256Method unknown—2/6/2026, 1:02:31 PM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 12:55:59 PM
Qwen3-8B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 12:49:40 PM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 12:43:43 PM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 12:37:08 PM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 12:31:10 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 12:24:40 PM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 12:18:22 PM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 12:11:58 PM
Qwen3-4B-AWQ ↗Text generationAWQvLLM256Method unknown—2/6/2026, 12:00:37 PM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM256Method unknown—2/6/2026, 9:45:17 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM256Method unknown—1/26/2026, 2:25:35 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM256Method unknown—1/25/2026, 4:49:37 PM
Qwen3-8B-AWQ ↗Text generationAWQvLLM256Method unknown—1/25/2026, 6:53:38 AM
Qwen3-4B-AWQ ↗Text generationAWQvLLM256Method unknown—1/25/2026, 6:17:50 AM
Gemma-3-270m ↗Text generationBF16vLLM256Method unknown—1/25/2026, 5:16:21 AM
gemma-3-270m ↗Text generationNativevLLM256Method unknown—1/25/2026, 4:14:45 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM256Method unknown—1/25/2026, 3:07:29 AM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM256Method unknown—1/23/2026, 4:21:33 AM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM256Method unknown—1/23/2026, 4:14:18 AM
Qwen3-32B-AWQ ↗Text generationAWQvLLM256Method unknown—1/23/2026, 3:17:48 AM
Qwen3-0.6B ↗Text generationBF16vLLM256Method unknown—1/22/2026, 10:04:49 PM
Qwen3-8B-AWQ ↗Text generationAWQvLLM256Method unknown—1/22/2026, 8:35:44 PM
Gemma-3-270m ↗Text generationBF16vLLM512Method unknown—3/6/2026, 10:01:36 PM
Gemma-3-270m ↗Text generationBF16vLLM512Method unknown—3/6/2026, 8:31:13 PM
Gemma-3-270m ↗Text generationBF16vLLM512Method unknown—3/6/2026, 8:02:55 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLM512Method unknown—2/7/2026, 2:28:53 AM
Qwen3-VL-32B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 4:57:04 PM
Qwen3-32B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 3:38:08 PM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 2:54:22 PM
Qwen3-30B-A3B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 2:08:49 PM
Qwen3-VL-30B-A3B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 12:52:05 PM
Qwen3-14B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 11:24:10 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 10:38:33 AM
Qwen3-8B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 9:27:35 AM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 8:35:09 AM
Qwen3-4B-Thinking-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 7:36:56 AM
Qwen3-VL-4B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 6:47:03 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 5:43:31 AM
Qwen3-4B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 5:16:27 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 4:54:16 AM
Qwen3-VL-2B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 4:31:39 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 4:01:58 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 3:43:19 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 3:17:13 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 3:00:09 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM512Method unknown—1/26/2026, 2:57:39 AM
Qwen3-1.7B-FP8 ↗Text generationFP8vLLM512Method unknown—1/25/2026, 5:01:45 PM
Qwen3-8B-AWQ ↗Text generationAWQvLLM512Method unknown—1/25/2026, 7:01:23 AM
Qwen3-4B-AWQ ↗Text generationAWQvLLM512Method unknown—1/25/2026, 6:24:57 AM
Gemma-3-270m ↗Text generationBF16vLLM512Method unknown—1/25/2026, 5:25:46 AM
gemma-3-270m ↗Text generationNativevLLM512Method unknown—1/25/2026, 4:23:12 AM
Qwen3-0.6B-FP8 ↗Text generationFP8vLLM512Method unknown—1/25/2026, 3:14:46 AM
Qwen3-VL-8B-FP8 ↗Text generationFP8vLLM512Method unknown—1/23/2026, 4:06:50 AM
Qwen3-4B-Instruct-FP8 ↗Text generationFP8vLLM512Method unknown—1/23/2026, 4:00:03 AM
Qwen3-30B-A3B-Thinking-FP8 ↗Text generationFP8vLLM512Method unknown—1/23/2026, 3:46:57 AM
Qwen3-14B-AWQ ↗Text generationAWQvLLM512Method unknown—1/23/2026, 3:09:48 AM
Qwen3.8-27B-NVFP4 ↗Text generationNVFP4vllmaime2026-test-zero-shot-strict-v4-n30-mt16384-t0quality-v163.33%10/15/2026, 9:07:41 AM
Qwen3.8-27B-NVFP4 ↗Text generationNVFP4vllmhle-test-self-judge-22ed307-v1-n5-mt16384-t0quality-v120%10/17/2026, 2:30:19 AM
Qwen3.8-27B-NVFP4 ↗Text generationNVFP4vllmifbench-test-official-loose-1c40f0c-v1-n3-mt16384-t0quality-v166.67%10/15/2026, 9:27:35 AM
Qwen3.8-27B-NVFP4 ↗Text generationNVFP4vllmifbench-test-official-loose-1c40f0c-v1-n300-mt16384-t0quality-v167.67%10/15/2026, 8:57:08 PM
Gemma-3-270m ↗Text generationBF16vLLMMixedMethod unknown—3/6/2026, 10:54:28 PM
Gemma-3-270m ↗Text generationBF16vLLMMixedMethod unknown—3/6/2026, 8:55:56 PM
Qwen3-4B-FP8 ↗Text generationFP8vLLMMixedMethod unknown—2/7/2026, 4:15:18 AM
Qwen3.8-27B-NVFP4 ↗Text generationNVFP4vllmmmlu-pro-test-zero-shot-strict-v4-n3-mt16384-t0quality-v166.67%10/15/2026, 6:20:58 AM
Qwen3.8-27B-NVFP4 ↗Text generationNVFP4vllmsimpleqa-verified-eval-self-judge-0dc97e0-v1-n5-mt16384-t0quality-v10%10/15/2026, 9:23:17 PM
Qwen3.8-27B-NVFP4 ↗Text generationNVFP4vllmsimpleqa-verified-eval-self-judge-0dc97e0-v1-n1000-mt16384-t0quality-v16.4%10/17/2026, 1:50:19 AM

Running benchmarks

0Updates automatically

No benchmarks are currently reported as running.

Methodology

Average of error-free runs for the selected recorded configuration. Missing methods and fallback runs are not ranked. Performance index = average / fastest matching hardware × 100. Value index = performance per purchase cost / best comparable value × 100. Quality is separate by dataset and protocol. Prices are dated seller listings, not completed transactions or market averages.

Updated 11/5/2026, 11:07:29 PM