New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the NVIDIA B200 run ERNIE-4.5-0.3B?
Yes — here's how fast and up to what context, with real measured numbers.
ERNIE-4.5-0.3B on the NVIDIA B200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2661k | 8111.1-9666.7 t/s | ~40.6-48.3 (slow) |
| Q8 | yes | 2667k | 16222.2-19333.3 t/s | ~81.1-96.7 (slow) |
| Q4 | yes | 2669k | 28968.3-34523.8 t/s | ~144.8-172.6 (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 B200 run ERNIE-4.5-0.3B?▶
Yes. At Q4 it generates ~28968.3-34523.8 tok/s and fits up to 2669k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~144.8-172.6 tok/s.
How much context fits?▶
At Q4, up to about 2669k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.