Qwen2.5-7B-Instruct

Qwen2.5-7B-Instruct

Qwen2.5-7B-Instruct is the 7-billion-parameter instruction-tuned variant of Alibaba Cloud's Qwen 2.5 model series, released under the Apache 2.0 license. It leads open-weight 7B models on multilingual reasoning, mathematics, and code generation, with a generous 128k token context window.

Key features

• 128k token context — full document and long-context reasoning
• Best-in-class 7B performance on MATH, HumanEval, and multilingual benchmarks
• Supports 29+ languages including Chinese, Japanese, Korean, Arabic
• Instruction-tuned with RLHF for helpful, harmless responses
• Available in GGUF/AWQ quantizations; Ollama-compatible

Quick start

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "Qwen/Qwen2.5-7B-Instruct"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

messages = [
    {"role": "system", "content": "You are a helpful coding assistant."},
    {"role": "user", "content": "Write a Python binary search function."}
]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer([text], return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(output[0], skip_special_tokens=True))

Install via ai-supply

npx ai-supply add qwen2-5-7b-instruct

Curated mirror of the open-source Qwen2.5-7B-Instruct (Apache-2.0). Get it from the source.