---
title: SmolVLM-Instruct-8bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/SmolVLM-Instruct-8bit"
md_url: "https://www.modelscope.cn/models/mlx-community/SmolVLM-Instruct-8bit.md"
repository: mlx-community/SmolVLM-Instruct-8bit
last_updated: 2025-02-06
license: apache-2.0
pipeline_tag: image-text-to-text
tasks:
  - image-text-to-text
model_type:
  - idefics3
architectures:
  - Idefics3ForConditionalGeneration
base_model:
  - HuggingFaceTB/SmolVLM-Instruct
base_model_relation: finetune
parameters: 728.8M
tensor_type:
  - F16
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
inference_backends:
  - "deploy_task vlm/text/emb"
  - "vllm 0.9.2"
downloads: 856
stars: 0
tags:
  - mlx
---

# SmolVLM-Instruct-8bit

> SmolVLM-Instruct-8bit - mlx-community 在 ModelScope 开源的模型。mlx-community/SmolVLM-Instruct-8bit This model was converted to MLX format from HuggingFaceTB/SmolVLM-Instruct using mlx-vlm version 0.1.2. Refer to the original model card for more details on the…

mlx-community/SmolVLM-Instruct-8bit 是 ModelScope 魔搭社区上的 728.8M 参数image-text-to-text模型，采用 apache-2.0 许可，基于 HuggingFaceTB/SmolVLM-Instruct 构建，可用 deploy_task vlm/text/emb、vllm 0.9.2 部署。

- **Repository**: mlx-community/SmolVLM-Instruct-8bit
- **License**: apache-2.0
- **Tasks**: image-text-to-text
- **Parameters**: 728.8M
- **Base model**: HuggingFaceTB/SmolVLM-Instruct
- **Inference backends**: deploy_task vlm/text/emb, vllm 0.9.2
- **Tags**: mlx
- **Downloads**: 856
- **Stars**: 0
- **Last updated**: 2025-02-06

Source: https://www.modelscope.cn/models/mlx-community/SmolVLM-Instruct-8bit

---

# mlx-community/SmolVLM-Instruct-8bit
This model was converted to MLX format from [`HuggingFaceTB/SmolVLM-Instruct`]() using mlx-vlm version **0.1.2**.
Refer to the [original model card](https://huggingface.co/HuggingFaceTB/SmolVLM-Instruct) for more details on the model.
## Use with mlx

```bash
pip install -U mlx-vlm
```

```bash
python -m mlx_vlm.generate --model mlx-community/SmolVLM-Instruct-8bit --max-tokens 100 --temp 0.0
```