---
title: Qwen3.8-27B-pi-mlx-6Bit
canonical_url: "https://www.modelscope.ai/models/SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit"
md_url: "https://www.modelscope.ai/models/SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit.md"
repository: SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit
last_updated: 2026-10-29
license: apache-2.0
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - qwen3_5
architectures:
  - Qwen3_5ForConditionalGeneration
base_model:
  - bytkim/Qwen3.8-27B-pi
base_model_relation: finetune
parameters: 5.9B
tensor_type:
  - BF16
  - U32
library_name:
  - mlx
  - safetensors
downloads: 6
stars: 0
tags:
  - mtp
  - multi-token-prediction
  - speculative-decoding
  - vision
  - image
  - multimodal
  - image-text-to-text
  - text-generation-inference
  - qwen
  - qwen3.8
  - qwen38
  - 27b
  - pi
  - coding
  - coder
  - code-generation
  - agent
  - tool-use
  - tool-calling
  - function-calling
  - reasoning
  - conversational
  - sft
  - grpo
  - reinforcement-learning
  - bf16
  - mlx
  - mlx-my-repo
---

# Qwen3.8-27B-pi-mlx-6Bit

> Qwen3.8-27B-pi-mlx-6Bit - An open-source model by SolsticeAI on ModelScope. SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit

SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit is a 5.9B-parameter text-generation model on ModelScope. licensed under apache-2.0. derived from bytkim/Qwen3.8-27B-pi.

- **Repository**: SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit
- **License**: apache-2.0
- **Tasks**: text-generation
- **Parameters**: 5.9B
- **Base model**: bytkim/Qwen3.8-27B-pi
- **Tags**: mtp, multi-token-prediction, speculative-decoding, vision, image, multimodal, image-text-to-text, text-generation-inference, qwen, qwen3.8, qwen38, 27b, pi, coding, coder, code-generation, agent, tool-use, tool-calling, function-calling, reasoning, conversational, sft, grpo, reinforcement-learning, bf16, mlx, mlx-my-repo
- **Downloads**: 6
- **Stars**: 0
- **Last updated**: 2026-10-29

Source: https://www.modelscope.ai/models/SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit

---

# SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit

The Model [SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit](https://huggingface.co/SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit) was converted to MLX format from [bytkim/Qwen3.8-27B-pi](https://huggingface.co/bytkim/Qwen3.8-27B-pi) using mlx-lm version **0.31.2**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("SolsticeAI/Qwen3.8-27B-pi-mlx-6Bit")

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)
```