L4 · 24 GB
Qwen3-4B-AWQ
Open a report for complete measurements.
- Task
- Text generation · Text → Text
- Workload
- 8K → 2K tokens · Prefix cache reuse
- Concurrency
- 4 concurrent requests
- Format · Engine
- AWQ · vllm · vLLM 0.29.0
- Unit
- tok/s
- Mean throughput
- 175tok/s
- Observed GPU occupancy
- 19.5 GiB
- First token
- 114 ms
- Reports combined
- 1
Launch this model
Open the measured report recipe to preserve its engine and MoE settings.
Use this recipeIndividual tests
Text → Text · Text generation · AWQ · vllm · 8192in-2048out · tok/s
Sortable table
Uncached prefill is unavailable when prompts reuse cache or reliable phase timings are missing. Open a report for its recorded phase timings and cache counts.
| Test date | Throughput | Concurrency / batch | P95 (s) | TTFT (ms) | Observed occupancy (GiB) | GPU power (W) | Decode (tok/s) | Uncached prefill (tok/s) | Parameters | View report |
|---|---|---|---|---|---|---|---|---|---|---|
| 10/23/2026, 6:48:39 AM | 175.39 tok/s | 4 | 46.748 | 114 | 19.55 | 71.93 | 43.94 | Uncached prefill unavailableView report | Parameters ↗ | 7e2cbfda ↗ |
Recorded test parameters
Configuration 1 · Parameters
- model
- Qwen/Qwen3-4B-AWQ
- engine
- vllm
- quant
- AWQ
- method
- wbench-fixed-http-v3
- task
- llm
- dataset
- migration-analysis-v1:sha256=68590c248e43f32c04dfd8062464fd1e4856917795fe307225e5c6112a54fe94:cache=prefix-reuse:repeats=3
- context
- 8192in-2048out
- input_tokens
- 8192
- output_tokens
- 2048
- duration
- 3600
- concurrency
- 4
- unit
- tok/s
- software
- { "kernel_selection": { "moe": { "activation_verification": "requested configuration and device compatibility; no per-layer dtype attestation", "input_quant_config": null, "preflight": null, "requested": { "humming_input_dtype": null, "moe_backend": "auto" }, "selection_messages": [], "verification": "unavailable" } }, "kv_capacity": { "gpu_cache_tokens": 119536, "maximum_concurrency": 11.12, "note": "Engine estimate, not measured saturation; hybrid cache token units may differ. Minimum reported worker bound.", "reference_tokens": 10752, "source": "vllm-startup-log" }, "packages": { "flash-attn": null, "flashinfer-python": "0.6.18", "sgl-kernel": null, "sglang": null, "tokenizers": "0.23.3", "torch": "2.13.0", "transformers": "5.20.2", "triton": "3.7.1", "vllm": "0.29.0" }, "python_implementation": "cpython", "python_version": "3.12.3", "schema": "wbench.python-runtime.v1", "scope": "launch-environment-installed-distributions", "startup": { "context": 10752, "max_num_seqs": 4, "source": "suite-startup-settings" } }
- engine_provenance
- { "checkpoint_quantization": { "mixed_quantization": false, "quantized_weight_formats": null, "recipe_label": "AWQ", "scope": "Declared quantized weight formats; unquantized modules and observed runtime kernels are separate evidence", "source": "checkpoint-metadata" }, "client_version": "1.1.57", "cpu_offload_gb": 0, "engine": "vllm", "gpu_driver_version": "580.82.07", "gpu_memory_utilization": 0.8999999761581421, "limit_mm_per_prompt": null, "model_config_sha256": "cf74d40352c502483cca3a94fd8037dce0d7b31c21831f1721ff6ee360b50075", "model_revision": "74d4bd2bd4bff9cafc9345221320bffb08b406a3", "reasoning_parser": "qwen3", "reported_cuda_version": "13.0", "runtime": { "kernel_selection": { "moe": { "activation_verification": "requested configuration and device compatibility; no per-layer dtype attestation", "input_quant_config": null, "preflight": null, "requested": { "humming_input_dtype": null, "moe_backend": "auto" }, "selection_messages": [], "verification": "unavailable" } }, "kv_capacity": { "gpu_cache_tokens": 119536, "maximum_concurrency": 11.12, "note": "Engine estimate, not measured saturation; hybrid cache token units may differ. Minimum reported worker bound.", "reference_tokens": 10752, "source": "vllm-startup-log" }, "packages": { "flash-attn": null, "flashinfer-python": "0.6.18", "sgl-kernel": null, "sglang": null, "tokenizers": "0.23.3", "torch": "2.13.0", "transformers": "5.20.2", "triton": "3.7.1", "vllm": "0.29.0" }, "python_implementation": "cpython", "python_version": "3.12.3", "schema": "wbench.python-runtime.v1", "scope": "launch-environment-installed-distributions", "startup": { "context": 10752, "max_num_seqs": 4, "source": "suite-startup-settings" } }, "shared_gpu": false }
- kernel_settings
- —
- precision
- —
- resolution
- —
- performance_task
- { "id": "migration-analysis-v1", "prompt_sha256": "fbb0d8755263f0d0fd563f991711ac6729c65cc92e78e2ab70c2e40e16ac2455", "sha256": "68590c248e43f32c04dfd8062464fd1e4856917795fe307225e5c6112a54fe94" }
- performance_scope
- —
- batch_size
- —
- dimensions
- —
- weight_quantization
- —
- max_sequence_length
- —
- model_scope
- —
- performance_dataset
- —
- workload_mode
- —
- graph_mode
- —
- model_revision
- —
- client_version
- 1.1.57