AutomationBench: Per-Task Model Costs Span an Order of Magnitude
ArtificialAnlys · x · 2026-07-07
The cost per task among models varies by over an order of magnitude: DeepSeek V4, Gemini 3.1 Flash-Lite, and Qwen3.7 Plus cost less than 5 cents, while Claude Opus 4.8(max) approaches $1.50. However, the leaders aren't always the most expensive: Gemini 3.5 Flash secured third place with a 42.6% score at just $0.49 per task.
Related event: AutomationBench-AA Launches to Evaluate AI Agents on SaaS Workflows(7 posts)→
More from Models
- Same Echo Maze prompt, three frontier models: all passed visually but shipped the same hidden bug — eyishazyer · 2026-09-11
- 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