Drift-Bench (Service-Oriented, No Clarification) - Intention: leaderboard

Metric: Success rate (%; 150 StableToolBench G1 API tasks with an intention fault (the stated goal changed), no clarification tools, mean of 3 trials). Source: arxiv.org. Saturation forecast: Around December 2026. 7 models tracked.

Top models

#ModelScore
1GLM-4.769.89
2DeepSeek V3.264.44
3Qwen 3 235B A22B 2507 Instruct57.22
4Llama 4 Maverick55.67
5GPT-5.254.33
6Gemini 2.5 Flash51
7GPT-OSS-120B42.56

Interactive version: theaggregate.ai/benchmark?slug=drift-bench-service-oriented-no-clarification-intention · How It Works · Data refreshed daily, snapshot 2026-09-25.