OmniBehavior - Live Streaming Binary Behaviors: leaderboard

OmniBehavior (2026) tests LLMs as simulators of real users: given the profile and long, real behaviour history of a user across the Kuaishou video, live-streaming, ads and e-commerce scenarios, the model predicts what the user does next. F1 on binary engagement decisions such as commenting in the live-streaming scenario, in percent; higher is better.

Metric: F1 (%) on binary engagement decisions (such as comment) in the live-streaming scenario, for predicting the real behaviour of 200 Kuaishou users (three months of interleaved video, live-stream, ads and e-commerce traces, 8,143 actions per user on average) from the user profile, history and current scenario, 6,000 balanced tasks, 32k context window, temperature 0.1; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Claude Opus 4.531.7#160
2GLM-4.728.97#380
3Claude Sonnet 428.86#410
4GPT-5.228.63#185
5Kimi K2 090528.6#570
6GPT-4o28.15#751
7OmniBehavior DeepSeek-V3 (checkpoint unspecified)27.92
8Claude Haiku 4.526.11#583
9Gemini 3 Flash25.61#175
10Claude Sonnet 4.525#261
11Qwen 3 235B A22B 2507 Instruct23.84#616

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

Interactive version: theaggregate.ai/benchmark?slug=omnibehavior-live-streaming-binary-behaviors · How It Works · Data refreshed daily, snapshot 2026-11-21.