LLM Inference Benchmark Explorer
Gemma-4-31B-it on RTX PRO 6000 — NVFP4, vLLM, TP1 inference benchmark
Open this configuration in the LLM Inference Benchmark Explorer
| Model | Gemma-4-31B-it |
|---|---|
| Parameters | 31B |
| Intelligence Index | 19 |
| Agentic Index | 4.2 |
| Device | RTX PRO 6000 |
| Quantization | NVFP4 |
| Max C | 32+ |
| Chat Capacity | 53 |
| Agentic Capacity | 18 |
| TP | — |
| DP | — |
| PP | — |
| Engine | vLLM |
| Speculative Decoding | — |
| C | TTFT (ms) | TPS (tok/s) | Status |
|---|---|---|---|
| 1 | 38 | 39.98 | PASS |
| 2 | 40 | 38.88 | PASS |
| 4 | 73 | 38.72 | PASS |
| 8 | 379 | 36.93 | PASS |
| 16 | 82 | 38.25 | PASS |
| 32 | 108 | 36.14 | PASS |
In OpenZeka's measurement, Gemma-4-31B-it (31B parameters), served in NVFP4 format with vLLM on RTX PRO 6000, reached a generation speed of 40 tok/s per request and a time to first token (TTFT) of 38 ms with a single request.
Considering the speed targets and the available KV cache capacity, the estimated capacity is 53 users for chat use and 18 for agentic use. The realistic capacity will likely fall between these two values. In scenarios dominated by coding, tool use, long workflows and multi-agent use, capacity approaches the agentic estimate; where shorter interactions, standard conversations and lighter tasks dominate, it approaches the chat estimate.
Intelligence Index and Agentic Index values are published by Artificial Analysis and are reproduced here with attribution.
In OpenZeka's measurement, Gemma-4-31B-it (31B parameters), served in NVFP4 format with vLLM on RTX PRO 6000, reached a generation speed of 40 tok/s per request and a time to first token (TTFT) of 38 ms with a single request.
Considering the speed targets and the available KV cache capacity, the estimated capacity is 53 users for chat use and 18 for agentic use. The realistic capacity will likely fall between these two values. In scenarios dominated by coding, tool use, long workflows and multi-agent use, capacity approaches the agentic estimate; where shorter interactions, standard conversations and lighter tasks dominate, it approaches the chat estimate.