MI

Mistral-Medium-3.5-128B

MistralGGUF by unslothUpdated 2026-05-02
Download on Hugging Face
Parameters125B
Context256K
Layers88
Architecturemistral3
Smallest quant34.9 GB
Quick answer

Mistral-Medium-3.5-128B needs about 78.3 GB of memory at Q4_K_M (a 74.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 24 quants, from 34.9 GB (IQ2_XXS) to 250.1 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 11.8 GB.

Mistral Medium 3.5 128B – dense 128B model with 256k context and vision support

Mistral Medium 3.5 128B is a dense 128B model with a 256k context window. It handles instruction following, reasoning, coding, and multimodal input, providing text output from both text and image inputs. The model supports function calls and configurable reasoning effort.

Good at

  • Large context reasoning
  • Multimodal text and image input
  • Function call and JSON output

Limits

  • Requires large VRAM for high‑quality quantizations
  • Vision support only in text output
  • License is Modified MIT with revenue restrictions

Best for

  • Complex agentic workflows
  • Long‑form document analysis
  • Multilingual instruction following

Who it’s for. Developers and researchers needing a unified model for instruction, reasoning, coding, and vision tasks. Suitable for applications that benefit from a single dense model with a large context window.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_XXS34.9 GB 38.4 GB
Too big Get file →
IQ2_M44.1 GB 47.6 GB
Too big Get file →
Q2_K46.6 GB 50.0 GB
Too big Get file →
Q2_K_L47.0 GB 50.4 GB
Too big Get file →
Q2_K_XL48.1 GB 51.6 GB
Too big Get file →
IQ3_XXS49.3 GB 52.7 GB
Too big Get file →
Q3_K_S54.4 GB (3 parts) 57.8 GB
Too big Get file →
Q3_K_M60.6 GB (3 parts) 64.1 GB
Too big Get file →
Q3_K_XL62.5 GB (3 parts) 66.0 GB
Too big Get file →
IQ4_XS67.1 GB (3 parts) 70.6 GB
Too big Get file →
IQ4_NL70.9 GB (3 parts) 74.3 GB
Too big Get file →
Q4_071.0 GB (3 parts) 74.5 GB
Too big Get file →
Q4_K_S71.2 GB (3 parts) 74.7 GB
Too big Get file →
Q4_K_M74.9 GB (3 parts) 78.3 GB
Too big Get file →
Q4_K_XL75.7 GB (3 parts) 79.1 GB
Too big Get file →
Q4_178.5 GB (3 parts) 81.9 GB
Too big Get file →
Q5_K_S86.2 GB (3 parts) 89.6 GB
Too big Get file →
Q5_K_M88.3 GB (3 parts) 91.8 GB
Too big Get file →
Q5_K_XL88.4 GB (3 parts) 91.8 GB
Too big Get file →
Q6_K102.6 GB (4 parts) 106.0 GB
Too big Get file →
Q6_K_XL109.2 GB (4 parts) 112.6 GB
Too big Get file →
Q8_0132.9 GB (4 parts) 136.3 GB
Too big Get file →
Q8_K_XL144.7 GB (4 parts) 148.2 GB
Too big Get file →
BF16250.1 GB (6 parts) 253.5 GB
Too big Get file →

For 8 GB VRAM, use IQ2_XXS (34.94 GB file, ~38.4 GB memory at 8k context). For 12 GB VRAM, Q4_K_S (71.25 GB file, ~74.7 GB memory) is a good balance. For 24 GB VRAM, Q4_K_XL (75.69 GB file, ~79.1 GB memory) offers higher quality.

What GPU do you need for Mistral-Medium-3.5-128B?

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: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: the best quant that fits is IQ2_M (44.1 GB file, about 47.6 GB in use).

Which Mistral-Medium-3.5-128B quant should you choose?

Q4_K_M is the usual starting point for Mistral-Medium-3.5-128B: it balances file size and quality well. Here it is a 74.9 GB file.

The Q8_0 file is 132.9 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.

If memory is tight, the smallest file is IQ2_XXS at 34.9 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

Architecturemistral3
Parameters125B
Layers88
Trained context262,144 tokens
KV cache (fp16)352 KB per token

Source

Quantized byunsloth
FamilyMistral
Licenseother
Downloads (HF)11,201
Last updated2026-05-02

Frequently asked questions about Mistral-Medium-3.5-128B

How much VRAM does Mistral-Medium-3.5-128B need?

About 78.3 GB at Q4_K_M with an 8K context window. That is the 74.9 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 Mistral-Medium-3.5-128B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 38.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Mistral-Medium-3.5-128B GGUF?

It depends on the quant: IQ2_XXS is 34.9 GB, IQ2_M is 44.1 GB, Q2_K is 46.6 GB, Q2_K_L is 47.0 GB, Q2_K_XL is 48.1 GB, IQ3_XXS is 49.3 GB and 18 more.

What context length does Mistral-Medium-3.5-128B support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 3.0 GB at 8K and 11.8 GB at 32K.

Where can I download Mistral-Medium-3.5-128B GGUF?

The files are on Hugging Face in the unsloth/Mistral-Medium-3.5-128B-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 Mistral-Medium-3.5-128B use?

Hugging Face lists the license as other. 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 Mistral Medium 3.5 on 8 GB VRAM?

Use the IQ2_XXS quantization; it requires about 38.4 GB memory at 8k context, which is beyond 8 GB VRAM, so it is not feasible without further optimization.

Does the model support image input?

Yes, it accepts image input and outputs text insights, but vision is only supported in text output.

What license applies to this model?

It is released under a Modified MIT License, allowing commercial use with revenue restrictions for large companies.

Other GGUFs of Mistral-Medium-3.5-128B

Similar models

Browse all →

More from unsloth

All by unsloth →

More Mistral models

All Mistral →

Other large models

All large models →