RTX 5090 · 32 GB
Qwen3.8-27B-NVFP4
Open a report for complete measurements.
- Task
- Text generation · Text → Text
- Workload
- 32K → 2K tokens
- Concurrency
- 1 request
- Format · Engine
- NVFP4 · vllm
- Unit
- tok/s
- Mean throughput
- 62.5tok/s
- Observed GPU occupancy
- 27.3 GiB
- First token
- 3.6 s
- Reports combined
- 1
Launch this model
The checkpoint precision is not recorded well enough to reproduce this configuration.
Individual tests
Text → Text · Text generation · NVFP4 · vllm · 32768in-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 |
|---|---|---|---|---|---|---|---|---|---|---|
| 10/14/2026, 11:14:30 PM | 62.48 tok/s | 1 | 32.792 | 3,553.5 | 27.29 | 433.29 | — | Uncached prefill unavailableView report | Parameters ↗ | 746bbc39 ↗ |
Recorded test parameters
Configuration 1 · Parameters
- model
- RadixArk/Qwen3.8-27B-NVFP4
- engine
- vllm
- quant
- NVFP4
- method
- wbench-fixed-http-v1
- task
- llm
- dataset
- synthetic-tokenizer-v1:repeats=2
- context
- 32768in-2048out
- input_tokens
- 32768
- output_tokens
- 2048
- duration
- 3600
- concurrency
- 1
- unit
- tok/s
- software
- —
- engine_provenance
- { "client_version": "1.1.10", "engine": "vllm" }
- kernel_settings
- —
- precision
- —
- resolution
- —
- performance_task
- —
- performance_scope
- —
- batch_size
- —
- dimensions
- —
- weight_quantization
- —
- max_sequence_length
- —
- model_scope
- —
- performance_dataset
- —
- workload_mode
- —
- graph_mode
- —
- model_revision
- —
- client_version
- 1.1.10