Speed test matrix
granite-4.2-8b-fp8 on RTX 3090 24GB
vllm · 0.32.0 · FP84 contexts (0–64K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K
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
- FP8
- 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 ibm-granite/granite-4.2-8b-fp8 --max-model-len 73728 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code
Prefill
prompt tok/s by context length8K context5,9711,380 ms TTFT
16K context5,1433,172 ms TTFT
32K context4,0458,027 ms TTFT
64K context2,81523,011 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 | 83.125 ms TTFT | 159.732 ms TTFT | 304.335 ms TTFT | 554.040 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity |
| 16K | 66.4106 ms TTFT | 111.695 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 32K | 55.4124 ms TTFT | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity | Over capacity |
| 64K | 41.4259 ms TTFT | Over capacity | 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 | 5,971.0 | — | 1,380 ms | — | — | 1 | — | — |
| Prefill | 16K | 1 | Valid | 5,143.0 | — | 3,172 ms | — | — | 1 | — | — |
| Prefill | 32K | 1 | Valid | 4,045.0 | — | 8,027 ms | — | — | 1 | — | — |
| Prefill | 64K | 1 | Valid | 2,815.0 | — | 23,011 ms | — | — | 1 | — | — |
| Sustained decode | 0 | 1 | Valid | 83.1 | 83.5 | 25 ms | 12.0 ms | 1.0 | 0 | 2,493 | 30.0 s |
| Sustained decode | 0 | 2 | Valid | 159.7 | 80.5 | 32 ms | 12.4 ms | 2.0 | 0 | 4,790 | 30.0 s |
| Sustained decode | 0 | 4 | Valid | 304.3 | 77.2 | 35 ms | 13.0 ms | 4.0 | 0 | 9,128 | 30.0 s |
| Sustained decode | 0 | 8 | Valid | 554.0 | 70.9 | 40 ms | 14.1 ms | 8.0 | 0 | 16,621 | 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 | 66.4 | 66.5 | 106 ms | 15.0 ms | 1.0 | 0 | 1,990 | 30.0 s |
| Sustained decode | 16K | 2 | Valid | 111.6 | 55.9 | 95 ms | 17.9 ms | 2.0 | 0 | 3,348 | 30.0 s |
| Sustained decode | 16K | 4 | Over capacity | — | — | — | — | — | — | — | — |
| 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 | 55.4 | 55.5 | 124 ms | 18.0 ms | 1.0 | 0 | 1,661 | 30.0 s |
| Sustained decode | 32K | 2 | Over capacity | — | — | — | — | — | — | — | — |
| 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 | 41.4 | 41.5 | 259 ms | 24.1 ms | 1.0 | 0 | 1,242 | 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 | — | — | — | — | — | — | — | — |
Over capacity: not run because this many requests at this context would exceed the server's KV cache.