CollabBench CWAH - Trustfulness: leaderboard
Affective collaboration quality in the CWAH-MultiPlayer household game of CollabBench (2026): trustfulness (reliable commitments and consistent, honest communication) of the evaluated agent, scored by a penalty-based LLM judge over windows of its reasoning, messages and actions, out of 5. Higher is better. The evaluated model plays one seat of a two-player game with a simulated partner whose personality varies.
Metric: Trustfulness score (out of 5): trustfulness (reliable commitments and consistent, honest communication), rated by a penalty-based DeepSeek-V3.1 judge: each window of three consecutive agent actions starts at 5 and loses 1 to 3 points per violation, and the trajectory score subtracts violation and worst-window terms and a message-sparsity penalty, clipped at 0; CWAH-MultiPlayer (Communicative Watch-And-Help household tasks with two embodied agents, 10 held-out test scenarios): the evaluated model drives Agent 1 inside the CoELA agent framework while a DeepSeek-V3.1 simulator role-plays a partner with a Big Five personality profile; mean over all evaluation trajectories; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 3.74 | #176 |
| 2 | Qwen 2.5 72B Instruct | 3.61 | #1092 |
| 3 | DeepSeek V3.1 | 3.35 | #497 |
| 4 | Qwen 2.5 7B Instruct | 2.58 | #2564 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=collabbench-cwah-trustfulness · How It Works · Data refreshed daily, snapshot 2026-11-13.