---
title: LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx
canonical_url: "https://www.modelscope.cn/models/nightmedia/LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx"
md_url: "https://www.modelscope.cn/models/nightmedia/LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx.md"
repository: nightmedia/LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx
last_updated: 2026-07-18
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - lfm2
architectures:
  - Lfm2ForCausalLM
parameters: 329.3M
tensor_type:
  - BF16
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
downloads: 355
stars: 0
tags:
  - mlx
---

# LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx

> LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx - nightmedia 在 ModelScope 开源的模型。LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx

- **Repository**: nightmedia/LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx
- **Tasks**: text-generation
- **Parameters**: 329.3M
- **Tags**: mlx
- **Downloads**: 355
- **Stars**: 0
- **Last updated**: 2026-07-18

Source: https://www.modelscope.cn/models/nightmedia/LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx

---

# LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("LFM2.5-1.2B-Thinking-Polaris-Claude-Deepseek-GLM-q8-mlx")

prompt = "hello"

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

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