Tencent · First MLX build 4 Jul 2025
Hunyuan-A13B-Instruct
MLX
Text tencent-hunyuan-a13b
A 80B parameter text generation model from Tencent, with 6 MLX builds, from 4-bit to bf16.
- Parameters
- 80B From Hugging Face
- Quantisation
- 4-bit Hunyuan-A13B-Instruct-4bit
- Memory needed
- 53 GB Estimate, unified memory
- Download
- 45.2 GB Weights, estimated
- Downloads
- 109 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 6 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 53 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 · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Hunyuan-A13B-Instruct-4bit --prompt "Hello"MLX builds
6 on mlx-community| Hunyuan-A13B-Instruct-4bit | 4-bit | 45.2 GB | 53 GB | 109 |
|---|---|---|---|---|
| Hunyuan-A13B-Instruct-4bit-DWQ | 4-bit DWQ | 45.2 GB | 53 GB | 28 |
| Hunyuan-A13B-Instruct-5bit | 5-bit | 55.3 GB | 64 GB | 33 |
| Hunyuan-A13B-Instruct-mixed-6-8bit | 8-bit | 85.4 GB | 98 GB | 36 |
| Hunyuan-A13B-Instruct-8bit | 8-bit | 85.4 GB | 98 GB | 26 |
| Hunyuan-A13B-Instruct-bf16 | bf16 | 161 GB | 182 GB | 19 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.