Qwen2.5-0.5B-Instruct is part of the Qwen2.5 series, featuring instruction-tuned language models with 0.5 billion parameters. Key highlights of this model include:
This model demonstrates significant improvements in instruction following, generating long texts, and understanding structured data. It supports multilingual capabilities across 29 languages, including English, Chinese, French, and more.
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
ttft 359.8ms
avg-token/s 10.32
apt install llm-model-qwen2.5-0.5b-prefill-20e
The Long-Context Model Compared to the Base Model, it provides extended context capabilities, offering a 256 context window and a maximum of 1,024 output tokens.
Support Platforms: LLM630 Compute Kit, Module LLM, and Module LLM Kit
256 context window
1,024 max output tokens
ttft 1126.19ms
avg-token/s 10.30
apt install llm-model-qwen2.5-0.5b-p256-ax630c
The INT4 Quantized Model compared to the Base Model, provides faster inference speed, offering a 128 context window and a maximum output of 1,024 tokens.
Support Platforms: LLM630 Compute Kit, Module LLM, Module LLM Kit
128 context window
1,024 max output tokens
ttft 442.95ms
avg-token/s 12.52
apt install llm-model-qwen2.5-0.5b-int4-ax630c