GSO-Bench leaderboard
- Claude Fable 5.1 leads
- 88.2%
- models measured
- 26
- lead over #2
- +8.8 pts
- median
- 18.6%
Leader on GSO-Bench · read October 21
# GSO-Bench, read October 21
leaders = {
"Claude Fable 5.1": 88.2%, "GPT-6 Astra": 79.4%, "Claude Fable 5": 78.4%,
}
On October 21, 2026, 29 of our catalog's models had a GSO-Bench result, and Anthropic Claude Fable 5.1 led with 88.2%, ahead of OpenAI GPT-6 Astra at 79.4%.
Software optimization: the model has to make real code run faster without breaking it. The number is the share of tasks where it matched the expert speed-up.
Role in the index
A coding board we show beside the Vibe Coding Index, not inside it: the index keeps the six boards it was defined on, so a score does not move every time a board is added. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Claude Fable 5.1Anthropic | 88.2% | 84.2 | $50.00 | |
| 2 | GPT-6 AstraOpenAI | 79.4% | 85.1 | $50.00 | |
| 3 | Claude Fable 5Anthropic | 78.4% | 79.3 | $50.00 | |
| 4 | GPT-5.6 SolOpenAI | 76.5% | 73.5 | $20.00 | |
| 5 | Claude Opus 4.8Anthropic | 47.1% | 65.0 | $25.00 | |
| 6 | Claude Opus 4.7Anthropic | 44.1% | 67.5 | $25.00 | |
| 7 | Claude Opus 4.6Anthropic | 41.2% | 57.2 | $25.00 | |
| 8 | GPT-5.5OpenAI | 40.2% | 71.3 | $30.00 | |
| 9 | Claude Sonnet 5Anthropic | 37.3% | 57.7 | $10.00 | |
| 10 | GPT-5.4OpenAI | 31.4% | 64.4 | $15.00 | |
| 11 | GPT-5.2OpenAI | 27.5% | 55.9 | $14.00 | |
| 12 | Claude Opus 4.5Anthropic | 26.5% | 44.0 | $25.00 | |
| 13 | Gemini 3.1 Pro PreviewGoogle | 22.6% | 32.3 | $12.00 | |
| 14 | Claude Sonnet 4.5Anthropic | 14.7% | 39.5 | $15.00 | |
| 15 | GPT-5.1OpenAI | 13.7% | 40.4 | $10.00 | |
| 16 | Gemini 3 Flash PreviewGoogle | 9.8% | 43.7 | $3.00 | |
| 17 | o3OpenAI | 8.8% | Not benchmarked | $8.00 | |
| 18 | Claude Opus 4RetiredAnthropic | 6.9% | Not benchmarked | $75.00 | |
| 19 | GPT-5OpenAI | 6.9% | 38.8 | $10.00 | |
| 20 | Kimi K2 0711Moonshot AI | 4.9% | Not benchmarked | $2.30 | |
| 21 | Qwen3 Coder 480B A35BQwen | 4.9% | Not benchmarked | $1.50 | |
| 22 | Claude 3.5 SonnetRetiredAnthropic | 4.6% | Not benchmarked | $15.00 | |
| 23 | Gemini 2.5 ProGoogle | 3.9% | 19.1 | $10.00 | |
| 24 | Gemini 2.5 Pro Preview 06-05Google | 3.9% | Not benchmarked | $10.00 | |
| 25 | Claude 3.7 SonnetRetiredAnthropic | 3.8% | Not benchmarked | $15.00 | |
| 26 | o4 MiniOpenAI | 3.6% | Not benchmarked | $4.40 | |
| 27 | GLM 4.5 AirZ.ai | 2.9% | Not benchmarked | $1.10 | |
| 28 | o3 MiniOpenAI | 1.3% | Not benchmarked | $4.40 | |
| 29 | GPT-4o (2024-11-20)OpenAI | 0.0% | Not benchmarked | $10.00 |
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.