ts-bench v1: leaderboard
A command-line coding agent driving the named model edits 25 Exercism TypeScript exercises until their test suites pass; the score is the share of the 25 solved in one run per agent and model. Aider Polyglot uses Exercism too, across six languages and with one fixed harness.
Metric: Success rate (%). Source: raw.githubusercontent.com. Saturation forecast: Estimated already saturated. 15 models tracked. Updated 17 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-5 | 96 | #177 | |
| 2 | Gemini 3 Pro (Preview) | 96 | #120 | |
| 3 | GPT-5.3 Codex | High | 96 | #122 |
| 4 | Gemini 2.5 Pro | 92 | #274 | |
| 5 | Gemini 3 Flash (Preview) | 92 | #151 | |
| 6 | GLM-4.6 | 92 | #496 | |
| 7 | GPT-5.2 | High | 92 | #185 |
| 8 | Qwen3 Coder Next | 92 | #724 | |
| 9 | Claude Sonnet 4.5 | 88 | #258 | |
| 10 | GPT-5.1 Codex | High | 88 | #284 |
| 11 | Gemini 2.5 Flash (Preview 09-2025) | 84 | #449 | |
| 12 | GLM-4.5 | 80 | #537 | |
| 13 | Qwen 3 Coder Plus (2025-09-23) | 76 | #708 | |
| 14 | Claude Sonnet 4 (20250514) | 72 | #405 | |
| 15 | Qwen 3 Coder Plus (2025-07-22) | 64 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=ts-bench-v1 · How It Works · Data refreshed daily, snapshot 2026-11-17.