Tele-AI · First MLX build 19 Jan 2026
TeleChat3-36B-Thinking
MLX
Text
A 36B parameter text generation model from Tele-AI, with 4 MLX builds, from 4-bit to bf16.
- Parameters
- 36B From Hugging Face
- Quantisation
- 4-bit TeleChat3-36B-Thinking-4bit
- Memory needed
- 25 GB Estimate, unified memory
- Download
- 20.4 GB Weights, estimated
- Downloads
- 32 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 4 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 25 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/TeleChat3-36B-Thinking-4bit --prompt "Hello"MLX builds
4 on mlx-community| TeleChat3-36B-Thinking-4bit | 4-bit | 20.4 GB | 25 GB | 32 |
|---|---|---|---|---|
| TeleChat3-36B-Thinking-6bit | 6-bit | 29.4 GB | 35 GB | 26 |
| TeleChat3-36B-Thinking-8bit | 8-bit | 38.5 GB | 46 GB | 31 |
| TeleChat3-36B-Thinking-bf16 | bf16 | 72.5 GB | 84 GB | 27 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.