Convex Coding Evals — No Guidelines (111 tasks): leaderboard
Convex Coding Evals — No Guidelines (111 tasks) evaluates models on writing Convex backends and client integrations without Convex-specific guidelines or documentation. The input modality is text and code. Its pinned metric is pass rate (%), averaged over the source leaderboard’s latest complete runs for this benchmark version. Source: BenchmarkList’s attributed result table.
Metric: Pass rate (%). Source: benchmarklist.com. Saturation forecast: Around February 2027. 32 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 84.68 | #6 |
| 2 | Claude Fable 5 | 83.78 | #20 |
| 3 | Claude Opus 5 | 81.68 | #16 |
| 4 | Claude Fable 5.1 | 81.08 | #10 |
| 5 | Claude Opus 4.8 | 80.63 | #39 |
| 6 | GPT-5.6 Sol | 80.48 | #22 |
| 7 | Grok 4.6 | 77.36 | #29 |
| 8 | Claude Sonnet 5 | 73.87 | #65 |
| 9 | Grok 4.5 | 72.97 | #54 |
| 10 | GPT-5.5 | 72.52 | #33 |
| 11 | Hy4 preview | 69.97 | #45 |
| 12 | GPT-5.6 Terra | 69.82 | #40 |
| 13 | Gemini 3.7 Flash | 68.65 | #24 |
| 14 | Claude Sonnet 4.6 | 68.47 | #94 |
| 15 | Kimi K3 | 65.77 | #35 |
No result here: #3 Claude Opus 5.5, #7 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=convex-coding-evals-no-guidelines-111-tasks · How It Works · Data refreshed daily, snapshot 2026-11-05.