Qwen2.5 72B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2: - Significantly more knowledge and has greatly improved capabilities in coding and...
Tested on 28 benchmarks · BenchGecko score 59.6. Top scores: Chatbot Arena Elo — Overall (1302.8%), ARC AI2 (92.7%), IFEval (86.4%).
DeepSeek V4 Flash 0731 scores 60.1 (101% as good) at $0.02/1M input · 96% cheaper
Code editing benchmark from the Aider project. Measures ability to apply targeted code changes while maintaining correctness and style.
Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.
BIG-Bench Hard. 23 challenging tasks from BIG-Bench where prior language models fell below average human performance.
HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.
Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.
HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.
Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.
- Typetext
- Context33K tokens (~16 books)
- ReleasedSep 2024
- LicenseOpen Source
- StatusActive
- Cost / Message~$0.001
Frequently Asked Questions
Key facts · as of 2026-10-05
- Qwen2.5 72B Instruct by Alibaba Qwen. BenchGecko score 59.6, rank 89 of 312 scored models (normalized average of public benchmark scores).
- List price $0.36 input · $0.40 output per 1M tokens (as of 2026-10-05).
- Sold by 2 providers (as of 2026-10-05): DeepInfra (fp8) $0.36 in / $0.40 out · Novita (bf16) $0.38 in / $0.40 out. Every provider
How to cite · data as of 2026-10-05
Qwen2.5 72B Instruct · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/qwen-2-5-72b-instruct
Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP