C3-Bench (Controllable Code Completion) - Implementation Control - Pass@1: leaderboard

Metric: Pass@1 (%; unit tests on the 1,286 instructed ICC tasks). Source: arxiv.org. Saturation forecast: Estimated already saturated. 42 models tracked.

Top models

#ModelScore
1O1 (2024-12-17)72.1
2Gemini 2.0 Flash70.7
3O1 Preview70.1
4Claude 3.5 Sonnet (20241022)68.8
5GPT-4o (2024-11-20)65.9
6GPT-4o (2024-08-06)65.8
7O1 Mini64.7
8DeepSeek V361.7
9DeepSeek V3 (0324)59.4
10Qwen 2.5 Coder 32B58.1
11Qwen 2.5 Coder 14B56.1
12GPT-4o (2024-05-13)54.3
13Claude 3.5 Haiku54.2
14Qwen 2.5 Coder 14B Instruct52
15Codestral-22B-v0.150.5

Interactive version: theaggregate.ai/benchmark?slug=c3-bench-controllable-code-completion-implementation-control-pass-1 · How It Works · Data refreshed daily, snapshot 2026-09-25.