Muse-Glimmer-30B-DFlash2 needs about 2.3 GB of memory at Q4_K_M (a 1.6 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 1.6 GB (Q4_K_M) to 5.6 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 0.7 GB.
Muse‑Glimmer‑30B‑DFlash2: a 2.77B parameter local LLM for agentic tasks
Muse‑Glimmer‑30B‑DFlash2 is a 2.77‑billion‑parameter causal language model optimized for local deployment. It supports multi‑step reasoning, reliable tool use, and multimodal input via a perception encoder. The model is released under Apache‑2.0 and runs with speculative decoding.
Good at
- Multi‑step reasoning
- Reliable tool use
- Multimodal input with images
Limits
- No vision support in the base model
- Requires speculative decoding server
- Large context requires 131072 tokens
Best for
- Building autonomous agents
- Complex code debugging
- Multimodal document analysis
Who it’s for. Developers and researchers who need a capable local LLM for agentic workflows and multimodal reasoning. The model is suitable for environments without cloud access.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 1.6 GB | 2.3 GB | Fits | Get file → | |
| Q8_0 | 3.0 GB | 3.6 GB | Fits | Get file → | |
| BF16 | 5.6 GB | 6.2 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_K_M (1.65 GB file, ~2.3 GB memory at 8 k context). For 16 GB VRAM, Q8_0 (2.96 GB file, ~3.6 GB memory). BF16 (5.56 GB file, ~6.2 GB memory) fits 32 GB VRAM setups. Estimates are approximate.
What GPU do you need for Muse-Glimmer-30B-DFlash2?
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: the best quant that fits is Q8_0 (3.0 GB file, about 3.6 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (3.0 GB file, about 3.6 GB in use).
- 8 GB VRAM: the best quant that fits is BF16 (5.6 GB file, about 6.2 GB in use).
Which Muse-Glimmer-30B-DFlash2 quant should you choose?
Q4_K_M is the usual starting point for Muse-Glimmer-30B-DFlash2: it balances file size and quality well. Here it is a 1.6 GB file.
The Q8_0 file is 3.0 GB, about 1.8 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.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Muse-Glimmer-30B-DFlash2
How much VRAM does Muse-Glimmer-30B-DFlash2 need?
About 2.3 GB at Q4_K_M with an 8K context window. That is the 1.6 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-DFlash2 on an 8 GB GPU?
Yes, with BF16: the 5.6 GB file needs about 6.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Muse-Glimmer-30B-DFlash2 GGUF?
It depends on the quant: Q4_K_M is 1.6 GB, Q8_0 is 3.0 GB, BF16 is 5.6 GB.
What context length does Muse-Glimmer-30B-DFlash2 support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.7 GB at 32K.
Where can I download Muse-Glimmer-30B-DFlash2 GGUF?
The files are on Hugging Face in the z-lab/Muse-Glimmer-30B-DFlash2-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-DFlash2 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‑DFlash2 on 8 GB VRAM?
Yes, the Q4_K_M quantization uses 1.65 GB file and requires about 2.3 GB VRAM at 8 k context, leaving headroom for KV cache.
Does the model support image input?
Yes, it includes a dedicated perception encoder for interleaved text and images.
What license applies to this model?
The model is released under the Apache‑2.0 license.