Speed test matrix
gpt-oss-safeguard-120b on RTX PRO 6000 Blackwell Server Edition 96GB
vllm · 0.30.1rc1.dev452+g73a583112 · MXFP45 contexts (8K–123K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 123K, 128K
Hardware
- Type
- Discrete GPU
- GPU
- RTX PRO 6000 Blackwell Server Edition
- VRAM
- 96 GB
- OS
- linux
Engine
- Engine
- vllm
- Version
- 0.30.1rc1.dev452+g73a583112
- Quantization
- MXFP4
- Backend
- cuda
- Repository
- github.com/vllm-project/vllm
- Build
- cuda-13.2
- Commit
73a5831127a9
Run
- Served model
- gpt-oss
- Revision
- main
- Benchmark tool
- llm-inference-bench v0.7.6
- Decode test
- Fixed duration
- Prefill mode
- Measured during decode warm-up
- Time per measurement
- 30 s
- Max output
- 4,096 tokens
Launch command
env MAX_JOBS=2 FLASHINFER_NVCC_THREADS=1 TORCHINDUCTOR_COMPILE_THREADS=1 "$HOME/env_rtxpro6k-openai-vllm-73a5831127a9/bin/vllm" serve openai/gpt-oss-safeguard-120b --served-model-name gpt-oss --trust-remote-code --tensor-parallel-size 1 --enable-auto-tool-choice --tool-call-parser openai --max-model-len 131072 --gpu-memory-utilization 0.82 --host 0.0.0.0 --port 8000 --api-key <redacted> --attention-backend FLASHINFER --moe-backend marlin
Prefill
prompt tok/s by context length8K context23.6K346 ms TTFT
16K context22.5K719 ms TTFT
32K context19.6K1,641 ms TTFT
64K context15.4K4,155 ms TTFT
123K context11K11,175 ms TTFT
128K context10.8K11,898 ms TTFT
Sustained decode
total tok/s · rows are context length, columns are simultaneous requests| Context | 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128 |
|---|---|---|---|---|---|---|---|---|
| 8K | 179.047 ms TTFT | 261.653 ms TTFT | 400.582 ms TTFT | 596.7100 ms TTFT | 831.4173 ms TTFT | Over capacity | Over capacity | Over capacity |
| 16K | 176.749 ms TTFT | 257.267 ms TTFT | 382.593 ms TTFT | 571.8161 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity |
| 32K | 165.058 ms TTFT | 240.987 ms TTFT | 377.4205 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 64K | 141.993 ms TTFT | 221.6162 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 123K | 116.3156 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
Over capacity: not run because this many requests at this context would exceed the server's KV cache.
TTFT is measured after the benchmark warms the server's prompt cache, so it mostly excludes prefill.
All measurements
Latencies are medians. Total tok/s is output across all simultaneous requests.| Phase | Context | Requests at once | Status | Total tok/s | Per request tok/s | TTFT | Time between tokens | Average active requests | Completed requests | Output tokens | Measured for |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Prefill | 8K | 1 | Valid | 23.6K | — | 346 ms | — | — | 1 | — | — |
| Prefill | 16K | 1 | Valid | 22.5K | — | 719 ms | — | — | 1 | — | — |
| Prefill | 32K | 1 | Valid | 19.6K | — | 1,641 ms | — | — | 1 | — | — |
| Prefill | 64K | 1 | Valid | 15.4K | — | 4,155 ms | — | — | 1 | — | — |
| Prefill | 123K | 1 | Valid | 11K | — | 11,175 ms | — | — | 1 | — | — |
| Prefill | 128K | 1 | Valid | 10.8K | — | 11,898 ms | — | — | 1 | — | — |
| Sustained decode | 8K | 1 | Valid | 179.0 | 179.8 | 47 ms | 5.6 ms | 1.0 | 1 | 5,040 | 28.2 s |
| Sustained decode | 8K | 2 | Valid | 261.6 | 134.0 | 53 ms | 7.5 ms | 2.0 | 2 | 7,846 | 30.0 s |
| Sustained decode | 8K | 4 | Valid | 400.5 | 101.5 | 82 ms | 9.9 ms | 4.0 | 0 | 12,014 | 30.0 s |
| Sustained decode | 8K | 8 | Valid | 596.7 | 77.2 | 100 ms | 12.9 ms | 8.0 | 0 | 17,899 | 30.0 s |
| Sustained decode | 8K | 16 | Valid | 831.4 | 55.1 | 173 ms | 18.2 ms | 16.0 | 0 | 24,937 | 30.0 s |
| Sustained decode | 8K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 8K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 8K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 16K | 1 | Valid | 176.7 | 173.0 | 49 ms | 5.8 ms | 1.0 | 1 | 5,300 | 30.0 s |
| Sustained decode | 16K | 2 | Valid | 257.2 | 130.7 | 67 ms | 7.7 ms | 2.0 | 2 | 7,716 | 30.0 s |
| Sustained decode | 16K | 4 | Valid | 382.5 | 99.5 | 93 ms | 10.1 ms | 4.0 | 0 | 11,473 | 30.0 s |
| Sustained decode | 16K | 8 | Valid | 571.8 | 73.7 | 161 ms | 13.6 ms | 8.0 | 0 | 17,148 | 30.0 s |
| Sustained decode | 16K | 16 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 16K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 16K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 16K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 32K | 1 | Valid | 165.0 | 161.4 | 58 ms | 6.2 ms | 1.0 | 1 | 4,951 | 30.0 s |
| Sustained decode | 32K | 2 | Valid | 240.9 | 121.1 | 87 ms | 8.3 ms | 2.0 | 0 | 7,224 | 30.0 s |
| Sustained decode | 32K | 4 | Valid | 377.4 | 95.6 | 205 ms | 10.5 ms | 4.0 | 0 | 11,320 | 30.0 s |
| Sustained decode | 32K | 8 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 32K | 16 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 32K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 32K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 32K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 1 | Valid | 141.9 | 143.1 | 93 ms | 7.0 ms | 1.0 | 1 | 4,257 | 30.0 s |
| Sustained decode | 64K | 2 | Valid | 221.6 | 111.2 | 162 ms | 9.0 ms | 2.0 | 0 | 6,648 | 30.0 s |
| Sustained decode | 64K | 4 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 8 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 16 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 1 | Valid | 116.3 | 116.6 | 156 ms | 8.6 ms | 1.0 | 0 | 3,461 | 29.8 s |
| Sustained decode | 123K | 2 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 4 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 8 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 16 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 128 | Over capacity | — | — | — | — | — | — | — | — |
Over capacity: not run because this many requests at this context would exceed the server's KV cache.