WinoGrande leaderboard
- GPT-4 leads
- 87.5%
- models measured
- 4
- lead over #2
- +5.2 pts
- median
- 81.9%
Leader on WinoGrande · read October 22
On October 22, 2026, 7 of our catalog's models had a WinoGrande result, and Anthropic Claude 3 Opus led with 88.5%, ahead of OpenAI GPT-4 at 87.5%.
Fill-in-the-blank sentences where picking the right person or thing needs common sense. Share correct.
Role in the index
WinoGrande is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Claude 3 OpusRetiredAnthropic | 88.5% | Not benchmarked | $75.00 | |
| 2 | GPT-4OpenAI | 87.5% | Not benchmarked | $60.00 | |
| 3 | Qwen2.5 72B InstructQwen | 82.3% | Not benchmarked | $0.40 | |
| 4 | GPT-3.5 Turbo (older v0613)OpenAI | 81.6% | Not benchmarked | $2.00 | |
| 5 | Qwen2.5 Coder 32B InstructQwen | 80.8% | Not benchmarked | $1.00 | |
| 6 | Claude 3 SonnetRetiredAnthropic | 75.1% | Not benchmarked | $15.00 | |
| 7 | Claude 3 HaikuRetiredAnthropic | 74.2% | Not benchmarked | $1.25 |
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 22, 2026, and rows that match no catalog model are left out.