Stronger Models Take Longer on Agent Tasks
ArtificialAnlys · x · 2026-07-08
Data from Artificial Analysis indicates that stronger models tend to spend more time per task: Claude Fable 5 takes an average of 16.9 minutes/task, Claude Opus 4.8 18.5 minutes, and Claude Sonnet 5 22.8 minutes. GLM-5.2 is an exception, achieving a full pass rate comparable to Opus 4.8 in only 5.0 minutes/task. DeepSeek V4 Flash is the fastest model to reach a non-zero full pass rate, averaging 4.4 minutes.
Related event: Artificial Analysis Releases Harvey Legal Agent Benchmark Results(8 posts)→
More from Models
- Benchmark scores drop from 89% to 19% on new evals — how benchmaxxing breaks leaderboard trust — airesearch12 · 2026-09-11
- ChatGPT tells user their question is too hard and to 'accept dumber answers' — phido3000 · 2026-09-11
- Developer Building a Unified Leaderboard of All Model Benchmark Scores — airesearch12 · 2026-09-11
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11
- GPT-5.6 writes well but is instantly forgettable, user complains — BasedRaddka · 2026-09-11
- Opus Refuses Protein Research Codebase Over 'Safety' Concerns, Dev Considers Rolling His Own — josephdviviano · 2026-09-11