🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN — FITS
Can the NVIDIA H200 run ChatGLM3-6B?
Yes — here's how fast and up to what context, with real measured numbers.
ChatGLM3-6B on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 887k | 292-348 t/s | ~2.4-2.9 (slow) |
| Q8 | yes | 932k | 584-696 t/s | ~4.9-5.8 (slow) |
| Q4 | yes | 953k | 1042.9-1242.9 t/s | ~8.7-10.4 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to system RAM offload (depends on your system RAM) and speed drops to the "beyond" figure.
FAQ
Can the NVIDIA H200 run ChatGLM3-6B?▶
Yes. At Q4 it generates ~1042.9-1242.9 tok/s and fits up to 953k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~8.7-10.4 tok/s.
How much context fits?▶
At Q4, up to about 953k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.