MI

Mixtral-instruct-56B

MistralGGUF by mradermacherUpdated 2024-08-02
Download on Hugging Face
Parameters46.7B
Context32K
Layers32
Architecturellama
Smallest quant9.8 GB
Quick answer

Mixtral-instruct-56B needs about 30.0 GB of memory at Q4_K_M (a 28.4 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 21 quants, from 9.8 GB (IQ1_S) to 38.4 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

Mixtral-instruct-56B: 46.7B parameter LLM with vision support

Mixtral-instruct-56B is a 46.7B parameter large language model based on the Mistral architecture. It supports vision tasks and is available in multiple GGUF quantizations. The model is licensed under apache-2.0.

Good at

  • Vision‑enabled instruction following
  • Large context handling up to 32768 tokens
  • Sparse Mixture of Experts architecture

Limits

  • No moderation mechanisms
  • Requires 8‑k context for memory estimates
  • No explicit language support list

Best for

  • Generating image‑caption pairs
  • Complex instruction following with visual input
  • Research on sparse expert models

Who it’s for. Developers and researchers needing a vision‑capable LLM with a large context window. Suitable for experimentation with sparse mixture of experts and for building applications that require both text and image understanding.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S9.8 GB 11.4 GB
Too big Get file →
IQ1_M10.8 GB 12.4 GB
Too big Get file →
IQ2_XXS12.6 GB 14.1 GB
Too big Get file →
IQ2_XS13.9 GB 15.5 GB
Too big Get file →
IQ2_S14.1 GB 15.7 GB
Too big Get file →
IQ2_M15.5 GB 17.1 GB
Too big Get file →
Q2_K17.3 GB 18.9 GB
Too big Get file →
IQ3_XXS18.2 GB 19.8 GB
Too big Get file →
IQ3_XS19.4 GB 20.9 GB
Too big Get file →
IQ3_S20.4 GB 22.0 GB
Too big Get file →
Q3_K_S20.4 GB 22.0 GB
Too big Get file →
IQ3_M21.4 GB 23.0 GB
Too big Get file →
Q3_K_M22.5 GB 24.1 GB
Too big Get file →
Q3_K_L24.2 GB 25.7 GB
Too big Get file →
IQ4_XS25.1 GB 26.7 GB
Too big Get file →
Q4_026.6 GB 28.1 GB
Too big Get file →
Q4_K_S26.7 GB 28.3 GB
Too big Get file →
Q4_K_M28.4 GB 30.0 GB
Too big Get file →
Q5_K_S32.2 GB 33.8 GB
Too big Get file →
Q5_K_M33.2 GB 34.8 GB
Too big Get file →
Q6_K38.4 GB 40.0 GB
Too big Get file →

For 8‑GB VRAM, use IQ1_S (9.82 GB file, 11.4 GB memory). For 16‑GB VRAM, IQ2_M (15.5 GB file, 17.1 GB memory) or Q3_K_M (22.55 GB file, 24.1 GB memory). For 24‑GB VRAM, Q4_K_M (28.45 GB file, 30 GB memory) or Q5_K_M (33.23 GB file, 34.8 GB memory). Memory estimates are approximate.

What GPU do you need for Mixtral-instruct-56B?

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 IQ1_S (9.8 GB file, about 11.4 GB in use).
  • 16 GB VRAM: the best quant that fits is IQ2_S (14.1 GB file, about 15.7 GB in use).
  • 24 GB VRAM: the best quant that fits is IQ3_M (21.4 GB file, about 23.0 GB in use).
  • 32 GB VRAM: the best quant that fits is Q4_K_M (28.4 GB file, about 30.0 GB in use).
  • 48 GB VRAM: the best quant that fits is Q6_K (38.4 GB file, about 40.0 GB in use).

Which Mixtral-instruct-56B quant should you choose?

Q4_K_M is the usual starting point for Mixtral-instruct-56B: it balances file size and quality well. Here it is a 28.4 GB file.

If memory is tight, the smallest file is IQ1_S at 9.8 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

Architecturellama
Parameters46.7B (approximate, mixture of experts)
Layers32
Trained context32,768 tokens
KV cache (fp16)128 KB per token

Source

Quantized bymradermacher
FamilyMistral
Licenseapache-2.0
Downloads (HF)4,336
Last updated2024-08-02

Frequently asked questions about Mixtral-instruct-56B

How much VRAM does Mixtral-instruct-56B need?

About 30.0 GB at Q4_K_M with an 8K context window. That is the 28.4 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 Mixtral-instruct-56B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 11.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Mixtral-instruct-56B GGUF?

It depends on the quant: IQ1_S is 9.8 GB, IQ1_M is 10.8 GB, IQ2_XXS is 12.6 GB, IQ2_XS is 13.9 GB, IQ2_S is 14.1 GB, IQ2_M is 15.5 GB and 15 more.

What context length does Mixtral-instruct-56B support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.

Where can I download Mixtral-instruct-56B GGUF?

The files are on Hugging Face in the mradermacher/Mixtral-instruct-56B-i1-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 Mixtral-instruct-56B 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 Mixtral-instruct-56B on 8 GB VRAM?

Use the IQ1_S quantization; it requires about 11.4 GB of VRAM at 8‑k context.

Does the model support vision tasks?

Yes, the model has detected vision abilities and can process image inputs.

What license applies to this model?

The model is released under the apache-2.0 license.

Other GGUFs of Mixtral-instruct-56B

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Mistral models

All Mistral →

Other large models

All large models →