InternScience · First MLX build 1 Jul 2026
Agents-A1
MLX
TextVision apache-2.0
A 35B parameter vision and text model from InternScience, with 7 MLX builds, from 3-bit to bf16.
- Parameters
- 35B From Hugging Face
- Quantisation
- 4-bit Agents-A1-4bit
- Memory needed
- 25 GB Estimate, unified memory
- Download
- 20.4 GB Weights, estimated
- Downloads
- 277 This build, last 30 days
- Task
- Vision and text Text, Vision
- MLX builds
- 7 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-vlm$ pip install -U mlx-vlm
$ python -m mlx_vlm.generate --model mlx-community/Agents-A1-4bit --image photo.jpg --prompt "Describe this image"MLX builds
7 on mlx-community| Agents-A1-3bit | 3-bit | 16.1 GB | 20 GB | 50 |
|---|---|---|---|---|
| Agents-A1-4bit | 4-bit | 20.4 GB | 25 GB | 277 |
| Agents-A1-OptiQ-4bit | 4-bit | 20.4 GB | 25 GB | 106 |
| Agents-A1-5bit | 5-bit | 24.7 GB | 30 GB | 53 |
| Agents-A1-6bit | 6-bit | 29.1 GB | 35 GB | 89 |
| Agents-A1-8bit | 8-bit | 37.7 GB | 45 GB | 117 |
| Agents-A1-bf16 | bf16 | 70.2 GB | 81 GB | 63 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.