MU

Muse Glimmer 30B Uncensored Heretic

MuseGGUF by mradermacherUpdated 2026-10-07
Download on Hugging Face
Parameters27.9B
Context128K
Layers52
Architecturemuse-glimmer
Smallest quant10.7 GB
Quick answer

Muse Glimmer 30B Uncensored Heretic needs about 17.9 GB of memory at Q4_K_M (a 16.9 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 11 quants, from 10.7 GB (Q2_K) to 29.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 1.7 GB.

Muse Glimmer 30B Uncensored Heretic – a 27.9B‑parameter multimodal LLM for local autonomous agents

Muse Glimmer 30B Uncensored Heretic is a 27.9‑billion‑parameter causal language model with a dedicated perception encoder. It supports vision and runs locally without cloud infrastructure. The model is released under Apache 2.0 and is intended for research and experimentation only.

Good at

  • Multimodal reasoning
  • Autonomous agentic task completion
  • Tool use and failure recovery

Limits

  • Uncensored content may be harmful or biased
  • Requires significant VRAM for larger quantizations
  • No official safety alignment

Best for

  • Research on agentic capabilities
  • Safety and alignment studies
  • Red‑team testing of multimodal LLMs

Who it’s for. Researchers, safety engineers, and developers who need a locally runnable multimodal LLM for experimentation. The model should not be deployed in public or end‑user services.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K10.7 GB 11.6 GB
Too big Get file →
Q3_K_S12.5 GB 13.4 GB
Too big Get file →
Q3_K_M13.7 GB 14.6 GB
Too big Get file →
Q3_K_L14.7 GB 15.6 GB
Too big Get file →
IQ4_XS15.3 GB 16.3 GB
Too big Get file →
Q4_K_S16.1 GB 17.1 GB
Too big Get file →
Q4_K_M16.9 GB 17.9 GB
Too big Get file →
Q5_K_S19.3 GB 20.3 GB
Too big Get file →
Q5_K_M19.8 GB 20.7 GB
Too big Get file →
Q6_K22.9 GB 23.8 GB
Too big Get file →
Q8_029.6 GB 30.5 GB
Too big Get file →

For 8 GB VRAM, use Q2_K (10.69 GB file, 11.6 GB memory at 8k context). For 12 GB VRAM, Q3_K_S (12.51 GB file, 13.4 GB memory). For 16 GB VRAM, Q4_K_M (16.94 GB file, 17.9 GB memory). Memory estimates are approximate.

What GPU do you need for Muse Glimmer 30B Uncensored Heretic?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: the best quant that fits is Q2_K (10.7 GB file, about 11.6 GB in use).
  • 16 GB VRAM: the best quant that fits is Q3_K_L (14.7 GB file, about 15.6 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (22.9 GB file, about 23.8 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (29.6 GB file, about 30.5 GB in use).

Which Muse Glimmer 30B Uncensored Heretic quant should you choose?

Q4_K_M is the usual starting point for Muse Glimmer 30B Uncensored Heretic: it balances file size and quality well. Here it is a 16.9 GB file.

The Q8_0 file is 29.6 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q2_K at 10.7 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturemuse-glimmer
Parameters27.9B
Layers52
Trained context131,072 tokens
KV cache (fp16)52 KB per token

Source

Quantized bymradermacher
FamilyMuse
Licenseapache-2.0
Downloads (HF)3,367
Last updated2026-10-07

Frequently asked questions about Muse Glimmer 30B Uncensored Heretic

How much VRAM does Muse Glimmer 30B Uncensored Heretic need?

About 17.9 GB at Q4_K_M with an 8K context window. That is the 16.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Muse Glimmer 30B Uncensored Heretic on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 11.6 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Muse Glimmer 30B Uncensored Heretic GGUF?

It depends on the quant: Q2_K is 10.7 GB, Q3_K_S is 12.5 GB, Q3_K_M is 13.7 GB, Q3_K_L is 14.7 GB, IQ4_XS is 15.3 GB, Q4_K_S is 16.1 GB and 5 more.

What context length does Muse Glimmer 30B Uncensored Heretic support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.7 GB at 32K.

Where can I download Muse Glimmer 30B Uncensored Heretic GGUF?

The files are on Hugging Face in the mradermacher/Muse-Glimmer-30B-Uncensored-Heretic-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Muse Glimmer 30B Uncensored Heretic use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Muse Glimmer 30B Uncensored Heretic on 8 GB VRAM?

Yes, the Q2_K quantization is 10.69 GB and requires about 11.6 GB VRAM at 8k context.

Is this model safe for production use?

No. The model has reduced safety alignment and is intended only for research and experimentation.

What license applies to this model?

The model is released under the Apache 2.0 license.

Other GGUFs of Muse-Glimmer-30B-Uncensored-Heretic

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Muse models

All Muse →

Other medium models

All medium models →