Devastaticglitch/Qwen2.5-72B-Instruct

TEXT GENERATIONPricing:Input $12 / Output $20Concurrent Unit Cost:4Model Size:72.7BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 18, 2026License:otherArchitecture:Transformer Featherless Exclusive Cold

Devastaticglitch/Qwen2.5-72B-Instruct is a 72.7 billion parameter instruction-tuned causal language model from the Qwen2.5 series, developed by Qwen. This model significantly enhances capabilities in coding, mathematics, and instruction following, while also improving long text generation and structured data understanding. It supports a full context length of 131,072 tokens and generation up to 8,192 tokens, with multilingual support for over 29 languages. It is particularly optimized for complex tasks requiring robust reasoning and structured output generation.

Loading preview...

Qwen2.5-72B-Instruct Overview

Qwen2.5-72B-Instruct is a 72.7 billion parameter instruction-tuned causal language model, part of the latest Qwen2.5 series. This model builds upon its predecessors with substantial improvements across several key areas, making it a versatile choice for demanding AI applications.

Key Capabilities

  • Enhanced Knowledge & Reasoning: Significantly improved performance in coding and mathematics, leveraging specialized expert models.
  • Superior Instruction Following: Demonstrates better adherence to instructions and is more resilient to diverse system prompts, aiding in role-play and conditional chatbot implementations.
  • Advanced Text Generation: Excels at generating long texts (over 8K tokens) and understanding/generating structured data, including JSON.
  • Extended Context & Multilingual Support: Features a full context length of 131,072 tokens (with YaRN for long texts) and supports generation up to 8,192 tokens. It also offers robust multilingual capabilities across more than 29 languages.

Good For

  • Complex Coding & Mathematical Tasks: Ideal for applications requiring high accuracy in programming and numerical reasoning.
  • Chatbots & Role-Playing: Its improved instruction following and system prompt resilience make it suitable for interactive agents.
  • Long Document Processing: Capable of handling and generating extensive textual content, beneficial for summarization or content creation from large inputs.
  • Structured Data Generation: Effective for tasks needing precise JSON or other structured output formats.

Finetunes
1 models