MI

midashenglm 0.6b

MidashenglmGGUF by mispeechUpdated 2026-04-17
Download on Hugging Face
Parameters596M
Context40K
Layers28
Architectureqwen3
Smallest quant0.3 GB
Quick answer

midashenglm 0.6b needs about 1.8 GB of memory at Q4_K_M (a 0.4 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 6 quants, from 0.3 GB (Q3_K_M) to 1.2 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 3.8 GB.

MidashengLM 0.6B – compact Qwen3‑based model with coding and vision abilities

MidashengLM 0.6B is a 0.596 billion‑parameter Qwen3 architecture model. It supports coding and vision tasks and is available in several GGUF quantizations. The model is released under the Apache‑2.0 license and can be run locally with llama.cpp.

Good at

  • Coding tasks
  • Vision inference
  • Audio captioning

Limits

  • Requires custom llama.cpp fork
  • No explicit language support listed
  • Limited context to 40960 tokens

Best for

  • Local inference on modest GPUs
  • Embedded systems with limited VRAM
  • Rapid prototyping of multimodal models

Who it’s for. Developers and researchers who need a lightweight multimodal model for local deployment. The model is suitable for those who prefer open‑source licensing and can work with the required llama.cpp fork.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_M0.3 GB 1.8 GB
Fits Get file →
Q4_K_M0.4 GB 1.8 GB
Fits Get file →
Q5_K_M0.4 GB 1.9 GB
Fits Get file →
Q6_K0.5 GB 1.9 GB
Fits Get file →
Q8_00.6 GB 2.1 GB
Fits Get file →
BF161.2 GB 2.6 GB
Fits Get file →

For 8 GB VRAM, Q4_K_M or Q3_K_M (0.4 GB or 0.35 GB) are recommended; both use about 1.8 GB memory at 8 k context. If more VRAM is available, Q5_K_M or Q6_K (0.44 GB or 0.5 GB) provide higher precision. BF16 (1.2 GB) offers the best quality but needs more memory.

What GPU do you need for midashenglm 0.6b?

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 BF16 (1.2 GB file, about 2.6 GB in use).

Which midashenglm 0.6b quant should you choose?

Q4_K_M is the usual starting point for midashenglm 0.6b: it balances file size and quality well. Here it is a 0.4 GB file.

The Q8_0 file is 0.6 GB, about 1.6 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 Q3_K_M at 0.3 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen3
Parameters596M
Layers28
Trained context40,960 tokens
KV cache (fp16)112 KB per token

Source

Quantized bymispeech
Licenseapache-2.0
Downloads (HF)2,957
Last updated2026-04-17

Frequently asked questions about midashenglm 0.6b

How much VRAM does midashenglm 0.6b need?

About 1.8 GB at Q4_K_M with an 8K context window. That is the 0.4 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 midashenglm 0.6b on an 8 GB GPU?

Yes, with BF16: the 1.2 GB file needs about 2.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of midashenglm 0.6b GGUF?

It depends on the quant: Q3_K_M is 0.3 GB, Q4_K_M is 0.4 GB, Q5_K_M is 0.4 GB, Q6_K is 0.5 GB, Q8_0 is 0.6 GB, BF16 is 1.2 GB.

What context length does midashenglm 0.6b support?

The GGUF header lists a trained context length of 40,960 tokens (40K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.8 GB at 32K.

Where can I download midashenglm 0.6b GGUF?

The files are on Hugging Face in the mispeech/midashenglm-0.6b-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 midashenglm 0.6b 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 MidashengLM 0.6B on 8 GB VRAM?

Yes, using Q4_K_M or Q3_K_M quantizations will fit within 8 GB VRAM, with an estimated 1.8 GB usage at 8 k context.

Does the model support multiple languages?

No language list is provided; the model’s language support is unspecified.

What is required to run this model?

A llama.cpp fork that supports the model’s architecture and quantizations is required for local inference.

Similar models

Browse all →

More from mispeech

All by mispeech →

Other tiny models

All tiny models →