AutomationBench-AA - Operations - Raw Objective Completion: leaderboard
Artificial Analysis's implementation of Zapier's AutomationBench restricted to its Operations tasks: the share of task objectives an agent completes in simulated SaaS applications, counted before the guardrail-violation adjustment used by the AutomationBench-AA headline. Reported 0-100 (%); higher is better.
Metric: Raw Objective Completion (%). Source: artificialanalysis.ai. Saturation forecast: Estimated already saturated. 243 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra (Max) | 97.01 | #6 (GPT-6 Astra) |
| 2 | Gemini 4 Carbon (High) | 97 | #1 (Gemini 4 Carbon) |
| 3 | GPT-6.1 Sol (Max) | 96.82 | #7 (GPT-6.1 Sol) |
| 4 | GPT-6 Astra (xHigh) | 96.57 | #6 (GPT-6 Astra) |
| 5 | GPT-6 Sol (xHigh) | 96.51 | #19 (GPT-6 Sol) |
| 6 | Gemini 4 Argon (High) | 96.39 | #9 (Gemini 4 Argon) |
| 7 | Muse Wonder Full (Max) | 96.21 | #12 (Muse Wonder Full) |
| 8 | Muse Wonder (Max) | 96.08 | #18 (Muse Wonder) |
| 9 | GPT-6.1 Sol (xHigh) | 95.94 | #7 (GPT-6.1 Sol) |
| 10 | GPT-6.1 Terra (Max) | 95.65 | #15 (GPT-6.1 Terra) |
| 11 | GPT-6 Sol (Max) | 95.63 | #19 (GPT-6 Sol) |
| 12 | GPT-5.6 Sol (Max) | 95.38 | #22 (GPT-5.6 Sol) |
| 13 | GPT-6 Astra (High) | 95.18 | #6 (GPT-6 Astra) |
| 14 | Claude Haiku 5.5 (Max) | 95.08 | #43 (Claude Haiku 5.5) |
| 15 | GPT-6 Sol (High) | 95.08 | #19 (GPT-6 Sol) |
No result here: #3 Claude Opus 5.5, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=automationbench-aa-operations-raw-objective-completion · How It Works · Data refreshed daily, snapshot 2026-11-05.