LLM Inference Benchmark Explorer
GLM-4.5-Air on DGX B300 — BF16, vLLM, TP1 inference benchmark
Open this configuration in the LLM Inference Benchmark Explorer
| Model | GLM-4.5-Air |
|---|---|
| Parameters | 110B |
| Intelligence Index | 11.1 |
| Agentic Index | — |
| Device | DGX B300 |
| Quantization | BF16 |
| Max C | 32+ |
| Chat Capacity | 16 |
| Agentic Capacity | 4 |
| TP | — |
| DP | — |
| PP | — |
| Engine | vLLM |
| Speculative Decoding | — |
| C | TTFT (ms) | TPS (tok/s) | Status |
|---|---|---|---|
| 1 | 62 | 107.87 | PASS |
| 2 | 40 | 76.61 | PASS |
| 4 | 44 | 65.40 | PASS |
| 8 | 58 | 50.75 | PASS |
| 16 | 116 | 38.78 | PASS |
| 32 | 372 | 28.44 | PASS |
In OpenZeka's measurement, GLM-4.5-Air (110B parameters), served in BF16 format with vLLM on DGX B300, reached a generation speed of 107.9 tok/s per request and a time to first token (TTFT) of 62 ms with a single request.
Considering the speed targets and the available KV cache capacity, the estimated capacity is 16 users for chat use and 4 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, GLM-4.5-Air (110B parameters), served in BF16 format with vLLM on DGX B300, reached a generation speed of 107.9 tok/s per request and a time to first token (TTFT) of 62 ms with a single request.
Considering the speed targets and the available KV cache capacity, the estimated capacity is 16 users for chat use and 4 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.