CodeSpecBench-Func - Completeness: leaderboard

CodeSpecBench (2026) asks models to write executable precondition and postcondition functions that specify the intended behaviour of a program; the function-level set uses 2,494 LeetCode problems with online-judge-validated valid and invalid tests. Completeness, the share of specifications that reject every invalid test, in percent; higher is better.

Metric: Completeness (%), the share of generated executable pre/postcondition specifications that reject every invalid test, on CodeSpecBench-Func (2,494 LeetCode problems; 50 valid and 50 invalid inputs plus valid and model-generated buggy outputs per problem, validated by the online judge); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 15 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-5 Mini59.7#364
2Gemini 2.5 Pro52.8#276
3GPT-552.4#178
4GPT-OSS-120B51.6#704
5GPT-OSS-20B51.4#1183
6Claude Sonnet 4.5Thinking46#261
7Gemini 2.5 Flash43.8#463
8QwQ-32B37.3#907
9Qwen 3 32B34.7#972
10DeepSeek V3.2Non-reasoning34.7#419
11Qwen 3 14B34.4#1269
12Qwen 3 8B25.7#1873
13Qwen 3 4B22.1#2348
14Qwen 3 1.7BNon-reasoning20.8#5860
15Qwen 3 0.6B17.2#8350

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=codespecbench-func-completeness · How It Works · Data refreshed daily, snapshot 2026-11-21.