Speed test matrix
Qwen3-8B-AWQ on RTX 3090 24GB
vllm · 0.32.0 · AWQ3 contexts (0–32K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K
Hardware
- Type
- Discrete GPU
- GPU
- RTX 3090
- VRAM
- 24 GB
- CPU
- Intel Core i7-9700K (8C)
- RAM
- 32 GB
- OS
- Ubuntu 24.04.5
Engine
- Engine
- vllm
- Version
- 0.32.0
- Quantization
- AWQ
- Backend
- cuda
- Repository
- github.com/vllm-project/vllm
- Build
- vllm/vllm-openai:v0.32.0-x86_64
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 RTX 3090 24GB (Zagitta/gpu-unlock), 420 W limit, PCIe Gen3 x16 | Intel Core i7-9700K 8C | 32 GB DDR4 | Ubuntu 24.04.5, NVIDIA driver 580.178.04 vllm serve Qwen/Qwen3-8B-AWQ --max-model-len 40960 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code
Prefill
prompt tok/s by context length8K context6,6601,239 ms TTFT
16K context5,6692,884 ms TTFT
32K context4,4347,340 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 | 126.725 ms TTFT | 237.722 ms TTFT | 436.728 ms TTFT | 751.742 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity |
| 16K | 94.863 ms TTFT | 151.989 ms TTFT | 260.4132 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 32K | 75.6131 ms TTFT | 110.9160 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 | 6,660.0 | — | 1,239 ms | — | — | 1 | — | — |
| Prefill | 16K | 1 | Valid | 5,669.0 | — | 2,884 ms | — | — | 1 | — | — |
| Prefill | 32K | 1 | Valid | 4,434.0 | — | 7,340 ms | — | — | 1 | — | — |
| Sustained decode | 0 | 1 | Valid | 126.7 | 128.0 | 25 ms | 7.8 ms | 1.0 | 0 | 3,801 | 30.0 s |
| Sustained decode | 0 | 2 | Valid | 237.7 | 120.9 | 22 ms | 8.3 ms | 2.0 | 0 | 7,130 | 30.0 s |
| Sustained decode | 0 | 4 | Valid | 436.7 | 112.2 | 28 ms | 8.9 ms | 4.0 | 0 | 13,100 | 30.0 s |
| Sustained decode | 0 | 8 | Valid | 751.7 | 98.3 | 42 ms | 10.2 ms | 8.0 | 0 | 22,553 | 30.0 s |
| Sustained decode | 0 | 16 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 0 | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 0 | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 0 | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 16K | 1 | Valid | 94.8 | 95.1 | 63 ms | 10.5 ms | 1.0 | 0 | 2,844 | 30.0 s |
| Sustained decode | 16K | 2 | Valid | 151.9 | 76.3 | 89 ms | 13.1 ms | 2.0 | 0 | 4,558 | 30.0 s |
| Sustained decode | 16K | 4 | Valid | 260.4 | 65.6 | 132 ms | 15.2 ms | 4.0 | 0 | 7,813 | 30.0 s |
| Sustained decode | 16K | 8 | Over capacity | — | — | — | — | — | — | — | — |
| 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 | 75.6 | 75.9 | 131 ms | 13.2 ms | 1.0 | 0 | 2,269 | 30.0 s |
| Sustained decode | 32K | 2 | Valid | 110.9 | 55.5 | 160 ms | 18.0 ms | 2.0 | 0 | 3,328 | 30.0 s |
| Sustained decode | 32K | 4 | Over capacity | — | — | — | — | — | — | — | — |
| 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 | — | — | — | — | — | — | — | — |
Over capacity: not run because this many requests at this context would exceed the server's KV cache.