9 models
Meta's latest open model built for always-on local agents. 30B parameters, licensed under Apache 2.0 and runs on a single GPU — tuned for tool use, long tasks, and failure recovery.
Reduced-refusal Muse Glimmer 30B via Heretic: 52/100 refusals (from 99), KL 0.063. Vision, Apache-2.0.
Unsloth UD Q3 Version of Muse Glimmer for a smaller footprint.
Q3 quantized variant of Muse Glimmer 30B for a slightly smaller footprint.
Imatrix-quantized GGUF builds of Meta's Muse Glimmer 30B (agentic, multimodal) down to 2-bit for 8–12GB setups — plus the official 17GB base build.
pulled from unsloth requant to IQ4_NL, with combined_en_huge as prompt (English + Coding focus)
A Muse Glimmer 30B fine-tune for Windows endpoint management: DSC v3, PowerShell 7, and Intune. Matches a dedicated domain fine-tune on accuracy while keeping full tool calling. Refuses legacy DSC 1.1 by design.
muse-glimmer:30b-dflash-256k-bf16
Runs on an RTX 3070 or better. I used mixed precision, IQ2, and CUDA optimizations.