CR

Neve Muse 5 12B

CrimsonGGUF by NeveAIUpdated 2026-08-10
Download on Hugging Face
Parameters12.2B
Context128K
Layers40
Architecturellama
Smallest quant13.0 GB
Quick answer

Neve Muse 5 12B needs about 14.9 GB of memory at Q8_0 (a 13.0 GB file) with an 8K context window, so it fits on a 16 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 5.4 GB.

Neve Muse 5 12B – 12.2B‑parameter conversational LLM

Neve Muse 5 12B is a 12.2B‑parameter llama architecture model optimized for human‑like conversation and creative writing. It is released under the Apache‑2.0 license and supports a 131072‑token context window.

Good at

  • Human‑like dialogue
  • Creative writing
  • Roleplay simulations

Limits

  • No vision support
  • Requires large VRAM
  • No explicit language list

Best for

  • Chatbot development
  • Creative content generation
  • Roleplay engines

Who it’s for. Developers building conversational agents or creative writing tools who need a large context window and permissive licensing. The model is suitable for applications that can allocate sufficient GPU memory.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_013.0 GB 14.9 GB
Too big Get file →

The only available quantization is Q8_0 with a file size of 13.02 GB. It requires approximately 14.9 GB of VRAM at 8 k context. For systems with 16 GB or more VRAM, Q8_0 is recommended. No other quants are provided.

What GPU do you need for Neve Muse 5 12B?

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: no quant fits at an 8K context.
  • 16 GB VRAM: the best quant that fits is Q8_0 (13.0 GB file, about 14.9 GB in use).

Which Neve Muse 5 12B quant should you choose?

Q8_0 (13.0 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters12.2B
Layers40
Trained context131,072 tokens
KV cache (fp16)160 KB per token

Source

Quantized byNeveAI
FamilyCrimson
Licenseapache-2.0
Downloads (HF)658
Last updated2026-08-10

Frequently asked questions about Neve Muse 5 12B

How much VRAM does Neve Muse 5 12B need?

About 14.9 GB at Q8_0 with an 8K context window. That is the 13.0 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 Neve Muse 5 12B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 14.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Neve Muse 5 12B GGUF?

It depends on the quant: Q8_0 is 13.0 GB.

What context length does Neve Muse 5 12B support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.3 GB at 8K and 5.4 GB at 32K.

Where can I download Neve Muse 5 12B GGUF?

The files are on Hugging Face in the NeveAI/Neve-Muse-5-12B-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 Neve Muse 5 12B 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 Neve Muse 5 12B on 8 GB VRAM?

No. The Q8_0 quant requires about 14.9 GB of VRAM at 8 k context, so 8 GB is insufficient.

Does Neve Muse 5 12B support vision tasks?

No, the model is purely text‑based and has no vision capabilities.

Other GGUFs of Crimson-Constellation-12B-heretic

Similar models

Browse all →

More from NeveAI

All by NeveAI →

More Crimson models

All Crimson →

Other medium models

All medium models →