GSM8K leaderboard
- models measured
- 6
- lead over #2
- +1.0 pts
- median
- 86.9%
Leader on GSM8K · read October 22
On October 22, 2026, 7 of our catalog's models had a GSM8K result, and Qwen2.5 Coder 32B Instruct led with 93.0%, ahead of OpenAI GPT-4 at 92.0%.
Grade-school word problems; most current models are near the ceiling. Share solved.
Role in the index
GSM8K is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Qwen2.5 Coder 32B InstructQwen | 93.0% | Not benchmarked | $1.00 | |
| 2 | GPT-4OpenAI | 92.0% | Not benchmarked | $60.00 | |
| 3 | GPT-4o-miniOpenAI | 91.3% | Not benchmarked | $0.60 | |
| 4 | Gemini 1.5 FlashRetiredGoogle | 82.4% | Not benchmarked | $0.30 | |
| 5 | Llama 3.1 8B InstructMeta Llama | 82.4% | Not benchmarked | $0.08 | |
| 6 | GPT-3.5 TurboOpenAI | 57.8% | Not benchmarked | $1.50 | |
| 7 | GPT-3.5 Turbo (older v0613)OpenAI | 46.9% | Not benchmarked | $2.00 |
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 22, 2026, and rows that match no catalog model are left out.