---
title: Falcon3-Mamba-7B-Instruct-4bits
canonical_url: "https://www.modelscope.cn/models/mlx-community/Falcon3-Mamba-7B-Instruct-4bits"
md_url: "https://www.modelscope.cn/models/mlx-community/Falcon3-Mamba-7B-Instruct-4bits.md"
repository: mlx-community/Falcon3-Mamba-7B-Instruct-4bits
last_updated: 2025-02-15
license: other
model_type:
  - falcon_mamba
architectures:
  - FalconMambaForCausalLM
base_model:
  - tiiuae/Falcon3-Mamba-7B-Instruct
base_model_relation: quantized
parameters: 1.1B
tensor_type:
  - U32
  - F16
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
inference_backends:
  - "deploy_task text/emb"
  - "vllm 0.9.2"
downloads: 1017
stars: 0
tags:
  - falcon3
  - falcon3_mamba
  - falcon_mamba
  - mlx
---

# Falcon3-Mamba-7B-Instruct-4bits

> Falcon3-Mamba-7B-Instruct-4bits - mlx-community 在 ModelScope 开源的模型。mlx-community/Falcon3-Mamba-7B-Instruct-4bits

mlx-community/Falcon3-Mamba-7B-Instruct-4bits 是 ModelScope 魔搭社区上的 1.1B 参数机器学习模型，采用 other 许可，基于 tiiuae/Falcon3-Mamba-7B-Instruct 构建，可用 deploy_task text/emb、vllm 0.9.2 部署。

- **Repository**: mlx-community/Falcon3-Mamba-7B-Instruct-4bits
- **License**: other
- **Parameters**: 1.1B
- **Base model**: tiiuae/Falcon3-Mamba-7B-Instruct
- **Inference backends**: deploy_task text/emb, vllm 0.9.2
- **Tags**: falcon3, falcon3_mamba, falcon_mamba, mlx
- **Downloads**: 1017
- **Stars**: 0
- **Last updated**: 2025-02-15

Source: https://www.modelscope.cn/models/mlx-community/Falcon3-Mamba-7B-Instruct-4bits

---

# mlx-community/Falcon3-Mamba-7B-Instruct-4bits

The Model [mlx-community/Falcon3-Mamba-7B-Instruct-4bits](https://huggingface.co/mlx-community/Falcon3-Mamba-7B-Instruct-4bits) was
converted to MLX format from [tiiuae/Falcon3-Mamba-7B-Instruct](https://huggingface.co/tiiuae/Falcon3-Mamba-7B-Instruct)
using mlx-lm version **0.20.2**.

## Use with mlx

```bash
pip install mlx-lm
```

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/Falcon3-Mamba-7B-Instruct-4bits")

prompt="hello"

if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
```