AgentCE-Bench - Workforce Scheduling: leaderboard

AgentCE-Bench (2026) tests tool-using agents on grid planning: the agent fills hidden slots of a weekly schedule by querying static JSON data, with decoy candidates that break global constraints. Task score in percent on the workforce scheduling domain; higher is better.

Metric: Task score (%) on AgentCE-Bench workforce domain (54 instances): the agent fills the hidden slots of a 5x7 planning grid with tool calls on static JSON (attribute filter queries, slot and global constraint checkers), each hidden slot having 25 candidates including decoys that violate only the global constraints; every cell is a whole count of instances over the 54 per domain (6 hidden-slot counts by 9 decoy budgets); at most 600 steps; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 13 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen 3.5 397B A17B90.7#270
2Qwen 3.5 122B A10B75.9#350
3Qwen 3.5 27B68.5#380
4MiniMax-M2.164.8#582
5Qwen 3.5 35B A3B61.1#494
6MiniMax-M2.561.1#600
7GLM-4.7 FP850#339
8Qwen 3.5 9B46.3#838
9MiniMax-M244.4#692
10Qwen 3.5 4B25.9#1270
11AgentCE-Bench MiroThinker-mini 30B (checkpoint unspecified)9.3
12Qwen 3.5 2B3.7#5380
13Qwen 3.5 0.8B0#8198

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

Interactive version: theaggregate.ai/benchmark?slug=agentce-bench-workforce-scheduling · How It Works · Data refreshed daily, snapshot 2026-11-22.