Speed test matrix
gemma-4-26B-A4B-it-NVFP4 on CMP 170HX 64GB
vllm · 0.32.0rc1.dev249+g18305948a · NVFP46 contexts (0–248K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 128K
Hardware
- Type
- Discrete GPU
- GPU
- CMP 170HX
- VRAM
- 64 GB
- CPU
- 2x Intel Xeon E5-2630 v3
- RAM
- 64 GB
- OS
- Ubuntu 26.04.1
Engine
- Engine
- vllm
- Version
- 0.32.0rc1.dev249+g18305948a
- Quantization
- NVFP4
- Backend
- cuda
- Repository
- github.com/vllm-project/vllm
- Build
- ghcr.io/morrowmake/vllm-cmp170hx@sha256:5b45c4d6feafefe2eba962375b0a9a0b64fd22b9e05abb4d1966f14a32a2db79
Run
- Served model
- bench
- Revision
- main
- Benchmark tool
- llm-inference-bench v0.8.0
- Decode test
- Fixed duration
- Prefill mode
- Measured during decode warm-up
- Time per measurement
- 30 s
- Max output
- 8,192 tokens
Launch command
# Untuned out-of-the-box baseline. Spec decode: none # 1x NVIDIA CMP 170HX 64GB (cmpunlocker), 180 W limit, PCIe Gen2 x4 | 2x Intel Xeon E5-2630 v3 | 64 GB DDR4 | Ubuntu 26.04.1, NVIDIA driver 610.57.04 | vLLM: Morrowmake vllm-cmp170hx image vllm serve RedHatAI/gemma-4-26B-A4B-it-NVFP4 --max-model-len 262144 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code --language-model-only
Prefill
prompt tok/s by context length8K context6,9951,139 ms TTFT
16K context5,7792,729 ms TTFT
32K context4,5006,977 ms TTFT
64K context3,02420,716 ms TTFT
128K context1,83068,396 ms TTFT
Sustained decode
total tok/s · rows are context length, columns are simultaneous requests| Context | 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128 |
|---|---|---|---|---|---|---|---|---|
| 0 | 149.473 ms TTFT | 264.4111 ms TTFT | 444.3107 ms TTFT | Looped | 1,031.8194 ms TTFT | 1,503.8361 ms TTFT | 2,071.8551 ms TTFT | 2,994.2716 ms TTFT |
| 16K | Looped | Looped | Looped | Looped | Looped | Looped | 1,336.24,064 ms TTFT | Over capacity |
| 32K | Looped | Looped | Looped | Looped | Looped | Looped | Over capacity | Over capacity |
| 64K | Looped | Looped | Looped | Looped | Looped | Over capacity | Over capacity | Over capacity |
| 128K | 73.21,231 ms TTFT | Looped | Looped | Looped | Over capacity | Over capacity | Over capacity | Over capacity |
| 248K | 50.4Requests queued2,439 ms TTFT | 90.03,427 ms TTFT | Looped | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
Requests queued: the server could not run all requests at once, so some waited in a queue. 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 | 6,995.0 | — | 1,139 ms | — | — | 1 | — | — |
| Prefill | 16K | 1 | Valid | 5,779.0 | — | 2,729 ms | — | — | 1 | — | — |
| Prefill | 32K | 1 | Valid | 4,500.0 | — | 6,977 ms | — | — | 1 | — | — |
| Prefill | 64K | 1 | Valid | 3,024.0 | — | 20,716 ms | — | — | 1 | — | — |
| Prefill | 128K | 1 | Valid | 1,830.0 | — | 68,396 ms | — | — | 1 | — | — |
| Sustained decode | 0 | 1 | Valid | 149.4 | 151.0 | 73 ms | 6.6 ms | 1.0 | 0 | 4,482 | 30.0 s |
| Sustained decode | 0 | 2 | Valid | 264.4 | 133.9 | 111 ms | 7.5 ms | 2.0 | 0 | 7,936 | 30.0 s |
| Sustained decode | 0 | 4 | Valid | 444.3 | 112.9 | 107 ms | 8.9 ms | 4.0 | 0 | 13,328 | 30.0 s |
| Sustained decode | 0 | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 0 | 16 | Valid | 1,031.8 | 66.9 | 194 ms | 14.9 ms | 16.0 | 0 | 30,960 | 30.0 s |
| Sustained decode | 0 | 32 | Valid | 1,503.8 | 49.6 | 361 ms | 20.2 ms | 32.0 | 0 | 45,120 | 30.0 s |
| Sustained decode | 0 | 64 | Valid | 2,071.8 | 34.3 | 551 ms | 29.1 ms | 64.0 | 0 | 62,144 | 30.0 s |
| Sustained decode | 0 | 128 | Valid | 2,994.2 | 24.9 | 716 ms | 40.1 ms | 128.0 | 0 | 89,984 | 30.1 s |
| Sustained decode | 16K | 1 | Looped | — | — | — | — | — | 0 | 0 | — |
| 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 | 1,336.2 | 21.2 | 4,064 ms | 47.1 ms | 64.0 | 0 | 40,064 | 30.0 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 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 32K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 1 | Looped | — | — | — | — | — | 0 | 0 | — |
| 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 | 128K | 1 | Valid | 73.2 | 73.7 | 1,231 ms | 13.6 ms | 1.0 | 0 | 2,195 | 30.0 s |
| Sustained decode | 128K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 128K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 128K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 128K | 16 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 128K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 128K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 128K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 248K | 1 | Requests queued | 50.4 | 53.8 | 2,439 ms | 18.6 ms | 0.9 | 0 | 1,511 | 30.0 s |
| Sustained decode | 248K | 2 | Valid | 90.0 | 45.6 | 3,427 ms | 21.9 ms | 2.0 | 0 | 2,700 | 30.0 s |
| Sustained decode | 248K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 248K | 8 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 248K | 16 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 248K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 248K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 248K | 128 | Over capacity | — | — | — | — | — | — | — | — |
Requests queued: the server could not run all requests at once, so some waited in a queue. 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.