Who leads the HumanEval+ benchmark?

Qwen2.5-Coder-32B currently leads HumanEval+ with a score of 87.2 on pass@1.

What is the state-of-the-art score on HumanEval+?

The state-of-the-art result on HumanEval+ is 87.2 (pass@1), achieved by Qwen2.5-Coder-32B as of 2026.

How many models are tracked on HumanEval+?

Codesota tracks 5 models on HumanEval+.

When was the HumanEval+ leaderboard last updated?

The HumanEval+ leaderboard on Codesota includes results through 2026.

Codesota · Benchmark · HumanEval+Home/Leaderboards/Code & Software Engineering/Code Generation/HumanEval+

Unknown

HumanEval+.

Name: HumanEval+ Benchmark Results
Creator: Unknown
Published: 2026-01-01
License: https://creativecommons.org/licenses/by/4.0/

Extended HumanEval with 80x more test cases. Tests code robustness and edge case handling.

Paper ↗Leaderboard ↓Lineage

§ 01 · Leaderboard

Results by metric.

Found a wrong score or missing run?

Use row edits to send a sourced correction into moderation.

Add / edit result ↗Report issue ↗

pass@1

Pass@1 is the reported evaluation metric for HumanEval+. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for pass@1verifiedpapervendorcommunityunverified

Muted rows were not state of the art when published — an earlier or same-year result already scored better.

Rank	Model	Trust	Score	Year	Links	Fix
01	Qwen2.5-Coder-32B Fetched from CodeSOTA API on 2026-04-20	verified	87.2	2026	Source ↗	Looks wrong?
02	DeepSeek-V3 Fetched from CodeSOTA API on 2026-04-20	verified	86.6	2026	Source ↗	Looks wrong?
03	GPT-4o Fetched from CodeSOTA API on 2026-04-20	verified	86	2026	Source ↗	Looks wrong?
04	DeepSeek-Coder-V2 Fetched from CodeSOTA API on 2026-04-20	verified	82.3	2026	Source ↗	Looks wrong?
05	DeepSeek-Coder-33B Fetched from CodeSOTA API on 2026-04-20	verified	75	2026	Source ↗	Looks wrong?