Unsloth Ships MiniMax-M3.1-Flash Quants: 2-bit Retains 91.5% Accuracy, Runs in 150GB RAM
danielhanchen · x · 2026-10-26
Unsloth released UD3 quants (GGUF) for MiniMax-M3.1-Flash, tested across multiple domains. The 2-bit quant retains 91.5% top-1 accuracy while being 45% smaller, running on 150GB RAM locally; UD-IQ1M retains 89.9% top-1.
By default the quants use NVFP4 for MoE layers and BF16/MXFP8 for the rest, and work directly in Unsloth Desktop. Unsloth calls the model among the strongest open models for its size.
More from Infra
- Revenue per dollar of compute up 5.2x in 5 quarters, undercutting 'massive lab losses' narrative — robleclerc · 2026-10-26
- $5M Seed Round Can't Afford One DeepSeek Training Run: GPU-less AI Startup's Three Paths — FinanceYF5 · 2026-10-26
- HBM demand shifts: low-stack inference plans shelved, Rubin to go 12-high only — BenBajarin · 2026-10-26
- CircuitHub unveils The Grid, a factory-scale robot making PCBs in the US twice as fast as Shenzhen — garrytan · 2026-10-26
- Ubicloud publishes "Fixing Tool-Call Accuracy in vLLM" — petrusenko_max · 2026-10-26
- Meta approves nonvoting Class C shares as 2026 capex guidance hits $135B–$145B — thione · 2026-10-26