🔌
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 GLM-10B-bidirectional?
Yes — here's how fast and up to what context, with real measured numbers.
GLM-10B-bidirectional on the FuriosaAI RNGD
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 138k | 48.8-59.3 t/s | ~1.3-1.6 (slow) |
| Q8 | yes | 199k | 97.5-118.5 t/s | ~2.6-3.2 (slow) |
| Q4 | yes | 226k | 174.1-211.6 t/s | ~4.6-5.6 (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 FuriosaAI RNGD run GLM-10B-bidirectional?▶
Yes. At Q4 it generates ~174.1-211.6 tok/s and fits up to 226k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~4.6-5.6 tok/s.
How much context fits?▶
At Q4, up to about 226k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.