Mistral AI · First MLX build 27 Jul 2026 · In Core AI since 8 Jun 2026
Mistral-7B-Instruct-v0.3
MLXCore AI
Text apache-2.0
A 7.2B parameter text generation model from Mistral AI, with one MLX build at 4-bit and Apple Core AI export recipes for macOS and iOS.
- Parameters
- 7.2B From Hugging Face
- Quantisation
- 4-bit Mistral-7B-Instruct-v0.3-mlx-4Bit
- Memory needed
- 7 GB Estimate, unified memory
- Download
- 4.1 GB Weights, estimated
- Downloads
- 91 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 1 On mlx-community
- Devices
- Mac Apple silicon
- Parameters
- 7.2B From Hugging Face
- Compression
- 4-bit, 4-bit weight palettized group8 Apple’s preset
- Memory needed
- 7 GB Estimate from parameters
- Context
- 8K tokens Fixed at export
- Precision
- FP16 Compute precision
- Platforms
- macOS, iOS Mac and iPhone
- Export presets
- 2 In Apple’s registry
- Listed by Apple
- 8 Jun 2026 First registry commit
Memory
Will it fit?Needs about 7 GB at 4-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/Mistral-7B-Instruct-v0.3-mlx-4BitMemory
Will it fit?Needs about 7 GB at 4-bit, 4-bit weight palettized group8
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the parameter count at Apple’s compression setting. Apple does not publish package sizes.
Export it
Terminal · uv$ git clone https://github.com/apple/coreai-models
$ cd coreai-models
$ uv run coreai.llm.export mistral-7b-instruct-v0.3
$ uv run coreai.llm.export mistral-7b-instruct-v0.3 --platform iOS
Apple publishes the recipe, not a ready-made package: these commands convert the original weights to a Core AI .aimodel on your Mac. See the repository for requirements.
MLX builds
1 on mlx-community| Mistral-7B-Instruct-v0.3-mlx-4Bit | 4-bit | 4.1 GB | 7 GB | 91 |
|---|
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.
Core AI export presets
From Apple’s registry| Preset | Platform | Compression | Precision | Context |
|---|---|---|---|---|
| mistral-7b-instruct-v0.3 | macOS | 4-bit | FP16 | 8K |
| mistral-7b-instruct-v0.3 | iOS | 4-bit weight palettized group8 | FP16 | 4K |