MLX Community · First MLX build 17 Jan 2024
Nous-Hermes-2-Mixtral-8x7B-DPO
MLX
apache-2.0
A 47B parameter model, with 2 MLX builds, from 8-bit to 4-bit.
- Parameters
- 47B From Hugging Face
- Quantisation
- 4-bit Nous-Hermes-2-Mixtral-8x7B-DPO-4bit
- Memory needed
- — GB Estimate, unified memory
- Download
- — Weights, estimated
- Downloads
- 657 This build, last 30 days
- Task
- —
- MLX builds
- 2 On mlx-community
- Devices
- Mac Apple silicon
Memory
Hugging Face does not list a weights size for this build, so there is no memory estimate.
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/Nous-Hermes-2-Mixtral-8x7B-DPO-4bitMLX builds
2 on mlx-community| Nous-Hermes-2-Mixtral-8x7B-DPO-8bit | 8-bit | 49.6 GB | 58 GB | 30 |
|---|---|---|---|---|
| Nous-Hermes-2-Mixtral-8x7B-DPO-4bit | 4-bit | — | — | 657 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.