Speed test matrix
Muse-Glimmer-30B-FP8-block on RTX PRO 6000 Blackwell Server Edition 96GB
vllm · 0.30.1rc1.dev452+g73a583112 · FP85 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
- FP8
- Backend
- cuda
- Repository
- github.com/vllm-project/vllm
- Build
- cuda-13.2
- Commit
73a5831127a9
Run
- Served model
- muse-glimmer
- 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-redhatai-vllm-73a5831127a9d2b8/bin/vllm" serve RedHatAI/Muse-Glimmer-30B-FP8-block --served-model-name muse-glimmer --tensor-parallel-size 1 --reasoning-parser muse_glimmer --enable-auto-tool-choice --tool-call-parser muse_glimmer --max-model-len 131072 --gpu-memory-utilization 0.86 --host 0.0.0.0 --port 8000 --api-key <redacted> --generation-config auto --speculative-config '{"method":"dflash","model":"meta-models/Muse-Glimmer-30B-assistant","num_speculative_tokens":15}'Prefill
prompt tok/s by context length8K context8,0731,015 ms TTFT
16K context7,9602,032 ms TTFT
32K context9,0363,560 ms TTFT
64K context8,4297,611 ms TTFT
123K context7,36316,720 ms TTFT
128K context7,37917,361 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 | Looped | Looped | Looped | Looped | Looped | Looped | 1,051.3902 ms TTFT | 1,113.72,070 ms TTFT |
| 16K | 172.3100 ms TTFT | Looped | Looped | Looped | Looped | Looped | 900.21,163 ms TTFT | Over capacity |
| 32K | Looped | Looped | Looped | Looped | Looped | 752.7996 ms TTFT | Over capacity | Over capacity |
| 64K | 128.5238 ms TTFT | Looped | Looped | Looped | Looped | Over capacity | Over capacity | Over capacity |
| 123K | 135.5319 ms TTFT | Looped | 464.4506 ms TTFT | Looped | Over capacity | Over capacity | Over capacity | Over capacity |
Looped: the model kept repeating the same text, so the benchmark discarded this measurement. 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 | 8,073.0 | — | 1,015 ms | — | — | 1 | — | — |
| Prefill | 16K | 1 | Valid | 7,960.0 | — | 2,032 ms | — | — | 1 | — | — |
| Prefill | 32K | 1 | Valid | 9,036.0 | — | 3,560 ms | — | — | 1 | — | — |
| Prefill | 64K | 1 | Valid | 8,429.0 | — | 7,611 ms | — | — | 1 | — | — |
| Prefill | 123K | 1 | Valid | 7,363.0 | — | 16,720 ms | — | — | 1 | — | — |
| Prefill | 128K | 1 | Valid | 7,379.0 | — | 17,361 ms | — | — | 1 | — | — |
| Sustained decode | 8K | 1 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 16 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 32 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 64 | Valid | 1,051.3 | 18.0 | 902 ms | 55.6 ms | 64.0 | 0 | 31,472 | 29.9 s |
| Sustained decode | 8K | 128 | Valid | 1,113.7 | 10.3 | 2,070 ms | 96.9 ms | 128.0 | 0 | 33,320 | 29.9 s |
| Sustained decode | 16K | 1 | Valid | 172.3 | 156.8 | 100 ms | 6.6 ms | 1.0 | 1 | 5,165 | 30.0 s |
| Sustained decode | 16K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 16 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 32 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 64 | Valid | 900.2 | 15.8 | 1,163 ms | 63.4 ms | 64.0 | 0 | 26,958 | 29.9 s |
| Sustained decode | 16K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 32K | 1 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 16 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 32 | Valid | 752.7 | 25.1 | 996 ms | 39.8 ms | 32.0 | 0 | 22,576 | 30.0 s |
| Sustained decode | 32K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 32K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 1 | Valid | 128.5 | 122.7 | 238 ms | 8.2 ms | 1.0 | 0 | 3,852 | 30.0 s |
| Sustained decode | 64K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 16 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 1 | Valid | 135.5 | 166.6 | 319 ms | 7.3 ms | 1.0 | 1 | 4,059 | 30.0 s |
| Sustained decode | 123K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 123K | 4 | Valid | 464.4 | 90.2 | 506 ms | 11.1 ms | 4.0 | 2 | 13,906 | 29.9 s |
| Sustained decode | 123K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 123K | 16 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 128 | Over capacity | — | — | — | — | — | — | — | — |
Looped: the model kept repeating the same text, so the benchmark discarded this measurement. Over capacity: not run because this many requests at this context would exceed the server's KV cache.