MLX Community · First MLX build 17 Jul 2024
DeepSeek-Coder-V2-Instruct-AQ4_1
MLX
deepseek-license
A 236B parameter model, with one MLX build.
- Parameters
- 236B From Hugging Face
- Quantisation
- Unstated DeepSeek-Coder-V2-Instruct-AQ4_1
- Memory needed
- 1056 GB Estimate, unified memory
- Download
- 943 GB Weights, estimated
- Downloads
- 333 This build, last 30 days
- Task
- —
- MLX builds
- 1 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 1056 GB at Unstated
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/DeepSeek-Coder-V2-Instruct-AQ4_1MLX builds
1 on mlx-community| DeepSeek-Coder-V2-Instruct-AQ4_1 | Unstated | 943 GB | 1056 GB | 333 |
|---|
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.