BigO(Bench) - Time Complexity Ranking: leaderboard

Each model, zero-shot, must write an optimised solution in a requested time-complexity class; the score is the percentile of the best measured complexity coefficient of 20 attempts among human solutions of that class (0 when no attempt passes the tests); each model ran on the 311 problems (640 problem-class pairs) of the test set. These are the original protocol's numbers (paper v1/v2, March 2025, problem statements with their input bounds); the paper's 2026 v3 drops the bounds and re-measures, and is not this board. Llama 3.1 405B, which generated the input parsers that decide which problems enter the test set, is not ranked. Unlike HumanEval-style correctness boards it scores whether code meets a stated algorithmic complexity.

Metric: coeffFull (%). Source: facebookresearch.github.io. Saturation forecast: Estimated already saturated. 16 models tracked. Updated 17 Nov 2026.

Top models

#ModelScoreOverall rank
1QwQ-32B50.4#908
2Llama 4 Maverick Instruct FP843.2#1022
3Llama 3.1 Nemotron Ultra 253B v141.6#1014
4Llama 4 Scout Instruct39#1575
5DeepSeek R1 Distill Qwen 32B38.6#1806
6DeepSeek R1 Distill Llama 70B38.3#1669
7Llama 3.3 70B Instruct33.8#1387
8DeepSeek V3 (0324)33.7#722
9DeepSeek V328.7#734
10Llama 3.1 70B Instruct28.3#1577
11DeepSeek-Coder-V2-Instruct-072427.7#1103
12openhands-lm-32B-v0.123.2#2237
13Codestral-22B-v0.121.6#2680
14Qwen 2.5 Coder 32B Instruct19.7#1597
15Llama 3.1 8B Instruct13.9#4576

No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=bigo-bench-time-complexity-ranking · How It Works · Data refreshed daily, snapshot 2026-11-17.