BigO(Bench) - Time Complexity Generation: leaderboard

Each model, zero-shot, must write a correct solution in each requested time-complexity class, checked by tests and the authors' profiling framework; all@1 counts a problem only when every class is met; each model ran on the 311 problems (640 problem-class pairs) of the test set, 20 samples each. 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: all@1 (%). Source: facebookresearch.github.io. Saturation forecast: Around May 2028. 16 models tracked. Updated 17 Nov 2026.

Top models

#ModelScoreOverall rank
1QwQ-32B9.6#908
2DeepSeek V3 (0324)7#722
3Llama 3.1 Nemotron Ultra 253B v16.1#1014
4Llama 4 Maverick Instruct FP85.3#1022
5DeepSeek R1 Distill Qwen 32B4.8#1806
6DeepSeek R1 Distill Llama 70B4.8#1669
7Llama 4 Scout Instruct3.5#1575
8DeepSeek V33.4#734
9Llama 3.3 70B Instruct3.3#1387
10DeepSeek-Coder-V2-Instruct-07243.3#1103
11Llama 3.1 70B Instruct3.1#1577
12Qwen 2.5 Coder 32B Instruct3.1#1597
13openhands-lm-32B-v0.12#2237
14Gemma 3 27B (IT)1.8#1339
15Codestral-22B-v0.11.3#2680

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-generation · How It Works · Data refreshed daily, snapshot 2026-11-17.