AI2 Reasoning Challenge leaderboard
- GPT-4 leads
- 96.3%
- models measured
- 3
- lead over #2
- +1.8 pts
- median
- 94.5%
Leader on AI2 Reasoning Challenge · read October 21
On October 21, 2026, 3 of our catalog's models had a AI2 Reasoning Challenge result, and OpenAI GPT-4 led with 96.3%, ahead of Qwen2.5 72B Instruct at 94.5%.
Grade-school science questions, the challenge set that simple retrieval cannot answer. Share answered correctly.
Role in the index
AI2 Reasoning Challenge is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | GPT-4OpenAI | 96.3% | Not benchmarked | $60.00 | |
| 2 | Qwen2.5 72B InstructQwen | 94.5% | Not benchmarked | $0.40 | |
| 3 | Qwen2.5 Coder 32B InstructQwen | 70.5% | Not benchmarked | $1.00 |
Only models in our catalog are listed, and each appears once with its best published run. 215 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out.