Speed test matrix
Phi-4-reasoning-plus-NVFP4 on RTX 3090 24GB
vllm · 0.32.0 · NVFP43 contexts (0–24K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 24K
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
- NVFP4
- 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 nvidia/Phi-4-reasoning-plus-NVFP4 --max-model-len 32768 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code
Prefill
prompt tok/s by context length8K context3,7882,229 ms TTFT
16K context3,3914,871 ms TTFT
24K context3,0608,035 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 | 82.940 ms TTFT | 162.630 ms TTFT | 313.738 ms TTFT | 587.444 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity |
| 16K | 71.754 ms TTFT | 139.682 ms TTFT | 240.6122 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 24K | 67.291 ms TTFT | 129.8104 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 | 3,788.0 | — | 2,229 ms | — | — | 1 | — | — |
| Prefill | 16K | 1 | Valid | 3,391.0 | — | 4,871 ms | — | — | 1 | — | — |
| Prefill | 24K | 1 | Valid | 3,060.0 | — | 8,035 ms | — | — | 1 | — | — |
| Sustained decode | 0 | 1 | Valid | 82.9 | 83.2 | 40 ms | 12.0 ms | 1.0 | 0 | 2,488 | 30.0 s |
| Sustained decode | 0 | 2 | Valid | 162.6 | 81.7 | 30 ms | 12.2 ms | 2.0 | 0 | 4,876 | 30.0 s |
| Sustained decode | 0 | 4 | Valid | 313.7 | 79.1 | 38 ms | 12.6 ms | 4.0 | 0 | 9,408 | 30.0 s |
| Sustained decode | 0 | 8 | Valid | 587.4 | 74.6 | 44 ms | 13.4 ms | 8.0 | 0 | 17,616 | 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 | 71.7 | 71.8 | 54 ms | 13.9 ms | 1.0 | 0 | 2,150 | 30.0 s |
| Sustained decode | 16K | 2 | Valid | 139.6 | 69.9 | 82 ms | 14.3 ms | 2.0 | 0 | 4,187 | 30.0 s |
| Sustained decode | 16K | 4 | Valid | 240.6 | 60.3 | 122 ms | 16.6 ms | 4.0 | 0 | 7,216 | 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 | 24K | 1 | Valid | 67.2 | 67.3 | 91 ms | 14.9 ms | 1.0 | 0 | 2,016 | 30.0 s |
| Sustained decode | 24K | 2 | Valid | 129.8 | 65.0 | 104 ms | 15.4 ms | 2.0 | 0 | 3,894 | 30.0 s |
| Sustained decode | 24K | 4 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 24K | 8 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 24K | 16 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 24K | 32 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 24K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 24K | 128 | Over capacity | — | — | — | — | — | — | — | — |
Over capacity: not run because this many requests at this context would exceed the server's KV cache.