Speed test matrix
Fara1.5-27B-AWQ-W4A16 on RTX 3090 24GB
vllm · 0.32.0 · AWQ W4A165 contexts (0–120K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 120K
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 W4A16
- 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. KV cache: int4_per_token_head # 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 Sohailhosseini/Fara1.5-27B-AWQ-W4A16 --max-model-len 131072 --gpu-memory-utilization 0.94 --max-num-seqs 52 --enable-prefix-caching --trust-remote-code --language-model-only --kv-cache-dtype int4_per_token_head --attention-backend TRITON_ATTN --max-cudagraph-capture-size 16
Prefill
prompt tok/s by context length8K context1,8084,531 ms TTFT
16K context1,49810,830 ms TTFT
32K context1,14128,313 ms TTFT
64K context771.083,578 ms TTFT
120K context493.0244,578 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 | 46.498 ms TTFT | 86.1101 ms TTFT | 166.0228 ms TTFT | 319.8390 ms TTFT | 366.7Requests queued578 ms TTFT | Over capacity | Over capacity | Over capacity |
| 16K | 41.2933 ms TTFT | Looped | 131.32,314 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 32K | 37.12,555 ms TTFT | 64.53,735 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 64K | 30.91,465 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 120K | 18.2Requests queued7,844 ms TTFT | Over capacity | Over capacity | 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 | 1,808.0 | — | 4,531 ms | — | — | 1 | — | — |
| Prefill | 16K | 1 | Valid | 1,498.0 | — | 10,830 ms | — | — | 1 | — | — |
| Prefill | 32K | 1 | Valid | 1,141.0 | — | 28,313 ms | — | — | 1 | — | — |
| Prefill | 64K | 1 | Valid | 771.0 | — | 83,578 ms | — | — | 1 | — | — |
| Prefill | 120K | 1 | Valid | 493.0 | — | 244,578 ms | — | — | 1 | — | — |
| Sustained decode | 0 | 1 | Valid | 46.4 | 46.4 | 98 ms | 21.6 ms | 1.0 | 0 | 1,390 | 30.0 s |
| Sustained decode | 0 | 2 | Valid | 86.1 | 43.1 | 101 ms | 23.2 ms | 2.0 | 0 | 2,581 | 30.0 s |
| Sustained decode | 0 | 4 | Valid | 166.0 | 41.6 | 228 ms | 24.1 ms | 4.0 | 0 | 4,976 | 30.0 s |
| Sustained decode | 0 | 8 | Valid | 319.8 | 40.1 | 390 ms | 24.9 ms | 8.0 | 0 | 9,593 | 30.0 s |
| Sustained decode | 0 | 16 | Requests queued | 366.7 | 32.9 | 578 ms | 30.4 ms | 12.1 | 0 | 10,926 | 29.8 s |
| Sustained decode | 0 | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 0 | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 0 | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 16K | 1 | Valid | 41.2 | 41.2 | 933 ms | 24.3 ms | 1.0 | 0 | 1,235 | 30.0 s |
| Sustained decode | 16K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 4 | Valid | 131.3 | 31.6 | 2,314 ms | 31.6 ms | 4.0 | 0 | 3,936 | 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 | 37.1 | 37.1 | 2,555 ms | 27.0 ms | 1.0 | 0 | 1,112 | 30.0 s |
| Sustained decode | 32K | 2 | Valid | 64.5 | 31.2 | 3,735 ms | 32.1 ms | 2.0 | 0 | 1,932 | 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 | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 1 | Valid | 30.9 | 31.0 | 1,465 ms | 32.3 ms | 1.0 | 0 | 928 | 30.0 s |
| Sustained decode | 64K | 2 | Over capacity | — | — | — | — | — | — | — | — |
| 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 | 120K | 1 | Requests queued | 18.2 | 24.0 | 7,844 ms | 41.6 ms | 0.7 | 0 | 545 | 30.0 s |
| Sustained decode | 120K | 2 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 120K | 4 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 120K | 8 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 120K | 16 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 120K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 120K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 120K | 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.