WildToolBench - Clarification Tasks: leaderboard

WildToolBench (2026) evaluates LLM tool use under realistic user behavior: 256 multi-turn scenarios over about 1,600 public APIs mix compositional tool orchestration, intentions hidden in earlier turns and switches between tool, chat and clarification requests. This board is accuracy in percent on tasks where the correct response is to ask the user for missing information; higher is better.

Metric: Accuracy (%) on tasks where the correct response is to ask the user for missing information, on WildToolBench (256 multi-turn scenarios, four user tasks each, 1,024 tasks, built from real user-log patterns with expert-annotated tool calls), each model through its native function-call format with default decoding; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 50 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Gemini 2.0 Flash (Thinking)52.34
2O148.05#407
3Gemini 2.5 Pro46.88#280
4O344.92#242
5GLM-4.544.53#540
6DeepSeek R143.75#484
7Claude Sonnet 441.8#414
8Claude Opus 4.141.8#284
9Claude 3.7 Sonnet41.41#455
10Qwen 3 30B A3BNon-reasoning41.41#1165
11Kimi K239.84#448
12Qwen 3 32BNon-reasoning39.84#983
13Qwen 3 8BThinking39.84#1890
14Qwen 3 14BNon-reasoning39.84#1284
15GPT-4o39.45#754

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

Interactive version: theaggregate.ai/benchmark?slug=wildtoolbench-clarification-tasks · How It Works · Data refreshed daily, snapshot 2026-11-22.