🔌
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 FuriosaAI RNGD run BTLM-3B?
Yes — here's how fast and up to what context, with real measured numbers.
BTLM-3B on the FuriosaAI RNGD
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 572k | 187.5-227.9 t/s | ~5-6.1 (slow) |
| Q8 | yes | 611k | 375-455.8 t/s | ~10-12.2 (slow) |
| Q4 | yes | 629k | 669.6-813.9 t/s | ~17.9-21.7 (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 FuriosaAI RNGD run BTLM-3B?▶
Yes. At Q4 it generates ~669.6-813.9 tok/s and fits up to 629k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~17.9-21.7 tok/s.
How much context fits?▶
At Q4, up to about 629k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.