Comet 60B (A12B) gets GGUF quants with MTP head for speculative decoding in llama.cpp
Felladrin · reddit · 2026-11-10
Redditor Felladrin converted Multiverse Computing's Comet 60B (A12B), a compressed Nemotron 3.5 Super, to GGUF so it runs on stock llama.cpp:
- Q4KM: 42.6 GB
- Q80: 64.2 GB
The quants keep the MTP head — pass --spec-type draft-mtp to llama-server for speculative decoding. Text-only for now; the vision tower wasn't converted.
More from Infra
- "Biblically accurate" low-loss waveguide crossing looks majestic, performs terribly — jwt0625 · 2026-11-10
- RBA gives LLaVA-OneVision-2 a 3x lossless training speed-up, merged upstream — liuziwei7 · 2026-11-10
- Nvidia lobbied hard against rules on remote access to restricted chips, officials say — rohanpaul_ai · 2026-11-10
- AI daily: Fable 5.2 tops evals, orchestrates 1000 agents for 8 days to build a prototype OS — jiqizhixin · 2026-11-10
- Engineer calls megakernel hype a psyop: CUDA graphs already express a strict superset — mike64_t · 2026-11-10
- Senate advances energy permitting bill 73-25, requiring data centers to pay more grid costs — firstadopter · 2026-11-10