---
license: apache-2.0
datasets:
- Azure99/blossom-v6.1-sft-stage1
- Azure99/blossom-v6.1-sft-stage2
language:
- zh
- en
base_model: Azure99/Blossom-V6.1-8b-128k
pipeline_tag: text-generation
tags:
- mlx
library_name: mlx
---

# Blossom-V6.1-8b-128k-q8-mlx

this is an experimental build to test long context

This model [Blossom-V6.1-8b-128k-q8-mlx](https://huggingface.co/Blossom-V6.1-8b-128k-q8-mlx) was
converted to MLX format from [Azure99/Blossom-V6.1-8b](https://huggingface.co/Azure99/Blossom-V6.1-8b)
using mlx-lm version **0.26.0**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("Blossom-V6.1-8b-128k-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
    )

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