MLX Community · First MLX build 14 Mar 2026
Qwen3-VL-Reranker-2B
MLX
Text apache-2.0
A 2.1B parameter reranking model, with 8 MLX builds, from 4-bit to bf16.
- Parameters
- 2.1B From Hugging Face
- Quantisation
- 4-bit Qwen3-VL-Reranker-2B-4bit
- Memory needed
- 4 GB Estimate, unified memory
- Download
- 1.8 GB Weights, estimated
- Downloads
- 163 This build, last 30 days
- Task
- Reranking Text
- MLX builds
- 8 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 4 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 · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/Qwen3-VL-Reranker-2B-4bitMLX builds
8 on mlx-community| Qwen3-VL-Reranker-2B-4bit | 4-bit | 1.8 GB | 4 GB | 163 |
|---|---|---|---|---|
| Qwen3-VL-Reranker-2B-mxfp4 | MXFP4 | 1.8 GB | 4 GB | 80 |
| Qwen3-VL-Reranker-2B-5bit | 5-bit | 2.0 GB | 5 GB | 79 |
| Qwen3-VL-Reranker-2B-6bit | 6-bit | 2.2 GB | 5 GB | 80 |
| Qwen3-VL-Reranker-2B-8bit | 8-bit | 2.6 GB | 5 GB | 151 |
| Qwen3-VL-Reranker-2B-mxfp8 | MXFP8 | 2.6 GB | 5 GB | 81 |
| Qwen3-VL-Reranker-2B-bf16 | bf16 | 4.3 GB | 7 GB | 95 |
| Qwen3-VL-Reranker-2B-nvfp4 | Unstated | 7.7 GB | 11 GB | 74 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.