🔌
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 Qwen2.5-1.5B?
Yes — here's how fast and up to what context, with real measured numbers.
Qwen2.5-1.5B on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1909k | 1137.7-1355.8 t/s | ~9.5-11.3 (slow) |
| Q8 | yes | 1933k | 2275.3-2711.7 t/s | ~19-22.6 (slow) |
| Q4 | yes | 1943k | 4063.1-4842.3 t/s | ~33.9-40.4 (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 Qwen2.5-1.5B?▶
Yes. At Q4 it generates ~4063.1-4842.3 tok/s and fits up to 1943k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~33.9-40.4 tok/s.
How much context fits?▶
At Q4, up to about 1943k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.