Speed test matrix
Qwen3-Next-80B-A3B-Thinking-AWQ-4bit on CMP 170HX 64GB
vllm · 0.32.0rc1.dev249+g18305948a · AWQ 4-bit6 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
- AWQ 4-bit
- 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 cyankiwi/Qwen3-Next-80B-A3B-Thinking-AWQ-4bit --max-model-len 262144 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code
Prefill
prompt tok/s by context length8K context7,0181,170 ms TTFT
16K context7,3882,199 ms TTFT
32K context6,6674,851 ms TTFT
64K context5,42811,889 ms TTFT
128K context3,91932,897 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 | 122.4145 ms TTFT | 217.7181 ms TTFT | 366.0257 ms TTFT | 592.8368 ms TTFT | 935.3413 ms TTFT | 1,423.5657 ms TTFT | 2,121.7902 ms TTFT | Over capacity |
| 16K | 116.6286 ms TTFT | 201.1361 ms TTFT | 326.5572 ms TTFT | 499.7925 ms TTFT | 730.91,554 ms TTFT | Over capacity | Over capacity | Over capacity |
| 32K | 110.8311 ms TTFT | 188.4399 ms TTFT | 296.3622 ms TTFT | 432.2967 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity |
| 64K | 103.0518 ms TTFT | 167.2684 ms TTFT | 249.51,034 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 128K | 90.01,003 ms TTFT | 135.21,198 ms TTFT | 188.11,858 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 248K | 72.11,746 ms TTFT | 99.42,236 ms TTFT | 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 | 7,018.0 | — | 1,170 ms | — | — | 1 | — | — |
| Prefill | 16K | 1 | Valid | 7,388.0 | — | 2,199 ms | — | — | 1 | — | — |
| Prefill | 32K | 1 | Valid | 6,667.0 | — | 4,851 ms | — | — | 1 | — | — |
| Prefill | 64K | 1 | Valid | 5,428.0 | — | 11,889 ms | — | — | 1 | — | — |
| Prefill | 128K | 1 | Valid | 3,919.0 | — | 32,897 ms | — | — | 1 | — | — |
| Sustained decode | 0 | 1 | Valid | 122.4 | 122.6 | 145 ms | 8.2 ms | 1.0 | 0 | 3,673 | 30.0 s |
| Sustained decode | 0 | 2 | Valid | 217.7 | 109.2 | 181 ms | 9.2 ms | 2.0 | 0 | 6,530 | 30.0 s |
| Sustained decode | 0 | 4 | Valid | 366.0 | 92.1 | 257 ms | 10.9 ms | 4.0 | 0 | 10,978 | 30.0 s |
| Sustained decode | 0 | 8 | Valid | 592.8 | 75.0 | 368 ms | 13.3 ms | 8.0 | 0 | 17,792 | 30.0 s |
| Sustained decode | 0 | 16 | Valid | 935.3 | 59.4 | 413 ms | 16.8 ms | 16.0 | 0 | 28,064 | 30.0 s |
| Sustained decode | 0 | 32 | Valid | 1,423.5 | 45.6 | 657 ms | 21.9 ms | 32.0 | 0 | 42,688 | 30.0 s |
| Sustained decode | 0 | 64 | Valid | 2,121.7 | 34.1 | 902 ms | 29.4 ms | 64.0 | 0 | 63,680 | 30.0 s |
| Sustained decode | 0 | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 16K | 1 | Valid | 116.6 | 116.8 | 286 ms | 8.6 ms | 1.0 | 0 | 3,498 | 30.0 s |
| Sustained decode | 16K | 2 | Valid | 201.1 | 100.6 | 361 ms | 9.9 ms | 2.0 | 0 | 6,032 | 30.0 s |
| Sustained decode | 16K | 4 | Valid | 326.5 | 81.6 | 572 ms | 12.2 ms | 4.0 | 0 | 9,796 | 30.0 s |
| Sustained decode | 16K | 8 | Valid | 499.7 | 62.1 | 925 ms | 16.1 ms | 8.0 | 0 | 14,984 | 30.0 s |
| Sustained decode | 16K | 16 | Valid | 730.9 | 45.1 | 1,554 ms | 22.2 ms | 16.0 | 0 | 21,936 | 30.0 s |
| Sustained decode | 16K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 16K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 16K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 32K | 1 | Valid | 110.8 | 111.0 | 311 ms | 9.0 ms | 1.0 | 0 | 3,323 | 30.0 s |
| Sustained decode | 32K | 2 | Valid | 188.4 | 94.4 | 399 ms | 10.6 ms | 2.0 | 0 | 5,652 | 30.0 s |
| Sustained decode | 32K | 4 | Valid | 296.3 | 74.2 | 622 ms | 13.5 ms | 4.0 | 0 | 8,888 | 30.0 s |
| Sustained decode | 32K | 8 | Valid | 432.2 | 53.9 | 967 ms | 18.5 ms | 8.0 | 0 | 12,960 | 30.0 s |
| 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 | 103.0 | 103.3 | 518 ms | 9.7 ms | 1.0 | 0 | 3,090 | 30.0 s |
| Sustained decode | 64K | 2 | Valid | 167.2 | 83.7 | 684 ms | 11.9 ms | 2.0 | 0 | 5,014 | 30.0 s |
| Sustained decode | 64K | 4 | Valid | 249.5 | 62.2 | 1,034 ms | 16.1 ms | 4.0 | 0 | 7,484 | 30.0 s |
| 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 | 128K | 1 | Valid | 90.0 | 90.3 | 1,003 ms | 11.1 ms | 1.0 | 0 | 2,698 | 30.0 s |
| Sustained decode | 128K | 2 | Valid | 135.2 | 67.6 | 1,198 ms | 14.8 ms | 2.0 | 0 | 4,056 | 30.0 s |
| Sustained decode | 128K | 4 | Valid | 188.1 | 46.7 | 1,858 ms | 21.4 ms | 4.0 | 0 | 5,644 | 30.0 s |
| Sustained decode | 128K | 8 | Over capacity | — | — | — | — | — | — | — | — |
| 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 | Valid | 72.1 | 72.6 | 1,746 ms | 13.8 ms | 1.0 | 0 | 2,164 | 30.0 s |
| Sustained decode | 248K | 2 | Valid | 99.4 | 49.9 | 2,236 ms | 20.0 ms | 2.0 | 0 | 2,980 | 30.0 s |
| Sustained decode | 248K | 4 | Over capacity | — | — | — | — | — | — | — | — |
| 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 | — | — | — | — | — | — | — | — |
Over capacity: not run because this many requests at this context would exceed the server's KV cache.