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
- AMD Ryzen 9 7900X (12C)
- 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), 350 W limit (VBIOS max), PCIe Gen4 x16 | AMD Ryzen 9 7900X 12C | 32 GB DDR5 | Ubuntu 24.04.5, NVIDIA driver 595.99.02 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 context1,5375,328 ms TTFT
16K context1,30812,405 ms TTFT
32K context1,00332,206 ms TTFT
64K context682.094,439 ms TTFT
120K context441.0273,785 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 | 45.371 ms TTFT | 84.1124 ms TTFT | 161.8215 ms TTFT | 312.4393 ms TTFT | 394.4Requests queued577 ms TTFT | Over capacity | Over capacity | Over capacity |
| 16K | Looped | Looped | 126.62,544 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 32K | 35.92,776 ms TTFT | 62.14,178 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 64K | 29.91,533 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 120K | 17.0Requests queued8,492 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 | 1,537.0 | — | 5,328 ms | — | — | 1 | — | — |
| Prefill | 16K | 1 | Valid | 1,308.0 | — | 12,405 ms | — | — | 1 | — | — |
| Prefill | 32K | 1 | Valid | 1,003.0 | — | 32,206 ms | — | — | 1 | — | — |
| Prefill | 64K | 1 | Valid | 682.0 | — | 94,439 ms | — | — | 1 | — | — |
| Prefill | 120K | 1 | Valid | 441.0 | — | 273,785 ms | — | — | 1 | — | — |
| Sustained decode | 0 | 1 | Valid | 45.3 | 45.4 | 71 ms | 22.0 ms | 1.0 | 0 | 1,358 | 30.0 s |
| Sustained decode | 0 | 2 | Valid | 84.1 | 42.1 | 124 ms | 23.7 ms | 2.0 | 0 | 2,524 | 30.0 s |
| Sustained decode | 0 | 4 | Valid | 161.8 | 40.5 | 215 ms | 24.7 ms | 4.0 | 0 | 4,852 | 30.0 s |
| Sustained decode | 0 | 8 | Valid | 312.4 | 39.2 | 393 ms | 25.5 ms | 8.0 | 0 | 9,368 | 30.0 s |
| Sustained decode | 0 | 16 | Requests queued | 394.4 | 33.6 | 577 ms | 29.8 ms | 12.6 | 0 | 11,737 | 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 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 4 | Valid | 126.6 | 30.3 | 2,544 ms | 33.0 ms | 4.0 | 0 | 3,796 | 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 | 35.9 | 35.9 | 2,776 ms | 27.8 ms | 1.0 | 0 | 1,077 | 30.0 s |
| Sustained decode | 32K | 2 | Valid | 62.1 | 29.9 | 4,178 ms | 33.5 ms | 2.0 | 0 | 1,863 | 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 | 29.9 | 29.9 | 1,533 ms | 33.5 ms | 1.0 | 0 | 896 | 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 | 17.0 | 23.2 | 8,492 ms | 43.1 ms | 0.7 | 0 | 511 | 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.