Kimi K3 Tech Report - Coding Experience (Codex): leaderboard
Moonshot AI's in-house Coding Experience evaluation measures the practical experience of working with a model as a coding agent in real development workflows; here GPT-5.6 Sol and GPT-5.5 run in OpenAI's Codex harness. Kimi K3 technical report, Table 3; score in percent, higher is better.
Metric: Score (%). Source: raw.githubusercontent.com. Saturation forecast: Around December 2027. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol (Max) | 59.3 | #22 (GPT-5.6 Sol) |
| 2 | GPT-5.5 (xHigh) | 56.8 | #33 (GPT-5.5) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=kimi-k3-tech-report-coding-experience-codex · How It Works · Data refreshed daily, snapshot 2026-11-05.