Qwen2.5-Coder-0.5B-Instruct

Introduction

Qwen2.5-Coder-0.5B-Instruct is Code-Specific Qwen large language models, Significantly improvements in code generation, code reasoning and code fixing. Key highlights of this model include:

• Type: Causal Language Model
• Training Stage: Pretraining & Post-training
• Architecture: Transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias, and tied word embeddings
• Number of Parameters: 0.49B (0.36B non-embedding)
• Number of Layers: 24
• Number of Attention Heads (GQA): 14 for Q and 2 for KV
• Context Length: Full 32,768 tokens and generation up to 8,192 tokens

Available NPU Models

Base Model

qwen2.5-Coder-0.5B-ax630c

The base model providing a 128 context window and a maximum output of 1,024 tokens.

Support Platforms: LLM630 Compute Kit, Module LLM, and Module LLM Kit

• 128 context window
• 1,024 max output tokens

Install

apt install llm-model-qwen2.5-coder-0.5b-ax630c

• Download llm-model-qwen2.5-coder-0.5b-ax630c