Speed test matrix
limite-1b-violetto on RTX PRO 6000 Blackwell Server Edition 96GB
vllm · 0.30.1rc1.dev452+g73a583112 · BF165 contexts (8K–123K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 123K, 128K
Hardware
- Type
- Discrete GPU
- GPU
- RTX PRO 6000 Blackwell Server Edition
- VRAM
- 96 GB
- OS
- linux
Engine
- Engine
- vllm
- Version
- 0.30.1rc1.dev452+g73a583112
- Quantization
- BF16
- Backend
- cuda
- Repository
- github.com/vllm-project/vllm
- Build
- cuda-13.2
- Commit
73a5831127a9
Run
- Served model
- limite-1b-violetto
- Revision
- main
- Benchmark tool
- llm-inference-bench v0.7.6
- Decode test
- Fixed duration
- Prefill mode
- Measured during decode warm-up
- Time per measurement
- 30 s
- Max output
- 4,096 tokens
Launch command
env VLLM_PLUGINS=limite MAX_JOBS=2 FLASHINFER_NVCC_THREADS=1 TORCHINDUCTOR_COMPILE_THREADS=1 "$HOME/env_rtxpro6k-paradigma-inc-vllm-73a5831127a9/bin/vllm" serve paradigma-inc/limite-1b-violetto --served-model-name limite-1b-violetto --tensor-parallel-size 1 --reasoning-parser deepseek_r1 --max-model-len 131072 --gpu-memory-utilization 0.83 --host 0.0.0.0 --port 8000 --api-key <redacted>
Prefill
prompt tok/s by context length8K context16.2K507 ms TTFT
16K context32K506 ms TTFT
32K context63.7K506 ms TTFT
64K context63.4K1,015 ms TTFT
123K context46.9K2,630 ms TTFT
128K context45.6K2,820 ms TTFT
Sustained decode
total tok/s · rows are context length, columns are simultaneous requests| Context | 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128 |
|---|---|---|---|---|---|---|---|---|
| 8K | 250.627 ms TTFT | Looped | Looped | Looped | Looped | Looped | Looped | Looped |
| 16K | 246.741 ms TTFT | 438.349 ms TTFT | Looped | Looped | Looped | Looped | Looped | Looped |
| 32K | Looped | Looped | Looped | Looped | Looped | Looped | Looped | Over capacity |
| 64K | Looped | Looped | Looped | Looped | Looped | Looped | Looped | Over capacity |
| 123K | 192.4215 ms TTFT | Looped | Looped | Looped | Looped | Looped | Over capacity | Over capacity |
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 | 16.2K | — | 507 ms | — | — | 1 | — | — |
| Prefill | 16K | 1 | Valid | 32K | — | 506 ms | — | — | 1 | — | — |
| Prefill | 32K | 1 | Valid | 63.7K | — | 506 ms | — | — | 1 | — | — |
| Prefill | 64K | 1 | Valid | 63.4K | — | 1,015 ms | — | — | 1 | — | — |
| Prefill | 123K | 1 | Valid | 46.9K | — | 2,630 ms | — | — | 1 | — | — |
| Prefill | 128K | 1 | Valid | 45.6K | — | 2,820 ms | — | — | 1 | — | — |
| Sustained decode | 8K | 1 | Valid | 250.6 | 250.9 | 27 ms | 4.0 ms | 1.0 | 2 | 7,517 | 30.0 s |
| Sustained decode | 8K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 16 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 32 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 64 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 8K | 128 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 1 | Valid | 246.7 | 247.6 | 41 ms | 4.0 ms | 1.0 | 2 | 7,401 | 30.0 s |
| Sustained decode | 16K | 2 | Valid | 438.3 | 219.6 | 49 ms | 4.6 ms | 2.0 | 2 | 13,148 | 30.0 s |
| Sustained decode | 16K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 16 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 32 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 64 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 16K | 128 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 1 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 16 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 32 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 64 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 32K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 64K | 1 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 16 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 32 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 64 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 64K | 128 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 1 | Valid | 192.4 | 194.0 | 215 ms | 5.2 ms | 1.0 | 1 | 5,773 | 30.0 s |
| Sustained decode | 123K | 2 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 123K | 4 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 123K | 8 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 123K | 16 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 123K | 32 | Looped | — | — | — | — | — | 0 | 0 | — |
| Sustained decode | 123K | 64 | Over capacity | — | — | — | — | — | — | — | — |
| Sustained decode | 123K | 128 | Over capacity | — | — | — | — | — | — | — | — |
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.