MLX Community · First MLX build 12 May 2024
dolphin-2.9.1-qwen-110b
MLX
other
A 111B parameter model, with 3 MLX builds, from 2-bit to 8-bit.
- Parameters
- 111B From Hugging Face
- Quantisation
- 2-bit dolphin-2.9.1-qwen-110b-2bit
- Memory needed
- 41 GB Estimate, unified memory
- Download
- 34.8 GB Weights, estimated
- Downloads
- 26 This build, last 30 days
- Task
- —
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 41 GB at 2-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/dolphin-2.9.1-qwen-110b-2bitMLX builds
3 on mlx-community| dolphin-2.9.1-qwen-110b-2bit | 2-bit | 34.8 GB | 41 GB | 26 |
|---|---|---|---|---|
| dolphin-2.9.1-qwen-110b-4bit | 4-bit | 62.6 GB | 72 GB | 23 |
| dolphin-2.9.1-qwen-110b-8bit | 8-bit | 118 GB | 135 GB | 17 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.