---
title: agentcpm-explore-mlx-fp16
canonical_url: "https://www.modelscope.cn/models/mlx-community/agentcpm-explore-mlx-fp16"
md_url: "https://www.modelscope.cn/models/mlx-community/agentcpm-explore-mlx-fp16.md"
repository: mlx-community/agentcpm-explore-mlx-fp16
last_updated: 2026-01-28
license: apache-2.0
model_type:
  - qwen3
architectures:
  - Qwen3ForCausalLM
base_model:
  - openbmb/AgentCPM-Explore
base_model_relation: finetune
parameters: 1.4B
tensor_type:
  - F16
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 29
stars: 0
tags:
  - mlx
---

# agentcpm-explore-mlx-fp16

> agentcpm-explore-mlx-fp16 - mlx-community 在 ModelScope 开源的模型。mstyslavity/AgentCPM-Explore-mlx-fp16

mlx-community/agentcpm-explore-mlx-fp16 是 ModelScope 魔搭社区上的 1.4B 参数机器学习模型，采用 apache-2.0 许可，基于 openbmb/AgentCPM-Explore 构建，可用 deploy_task text/emb、lmdeploy_turbomind 0.9.1、sglang 0.5.2 部署。

- **Repository**: mlx-community/agentcpm-explore-mlx-fp16
- **License**: apache-2.0
- **Parameters**: 1.4B
- **Base model**: openbmb/AgentCPM-Explore
- **Inference backends**: deploy_task text/emb, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2
- **Tags**: mlx
- **Downloads**: 29
- **Stars**: 0
- **Last updated**: 2026-01-28

Source: https://www.modelscope.cn/models/mlx-community/agentcpm-explore-mlx-fp16

---

# mstyslavity/AgentCPM-Explore-mlx-fp16

The Model [mstyslavity/AgentCPM-Explore-mlx-fp16](https://huggingface.co/mstyslavity/AgentCPM-Explore-mlx-fp16) was converted to MLX format from [openbmb/AgentCPM-Explore](https://huggingface.co/openbmb/AgentCPM-Explore) using mlx-lm version **0.29.1**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("mstyslavity/AgentCPM-Explore-mlx-fp16")

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)
```