AutomationBench-AA - Sales - Raw Objective Completion: leaderboard
Artificial Analysis's implementation of Zapier's AutomationBench restricted to its Sales 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: Around February 2027. 243 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Carbon (High) | 92.21 | #1 (Gemini 4 Carbon) |
| 2 | Gemini 4 Argon (High) | 92 | #9 (Gemini 4 Argon) |
| 3 | Kimi K3.1 Pro | 91.83 | #17 |
| 4 | GPT-6 Sol (xHigh) | 90.07 | #19 (GPT-6 Sol) |
| 5 | GPT-6 Sol (Max) | 88.68 | #19 (GPT-6 Sol) |
| 6 | Muse Wonder Full (Max) | 88.19 | #12 (Muse Wonder Full) |
| 7 | DeepSeek V4.1 Flash (1017) (High) | 87.73 | #27 (DeepSeek V4.1 Flash (1017)) |
| 8 | Grok 4.6 (xHigh) | 87.62 | #29 (Grok 4.6) |
| 9 | Claude Opus 5 (Max) | 86.9 | #16 (Claude Opus 5) |
| 10 | GPT-6 Sol (High) | 86.84 | #19 (GPT-6 Sol) |
| 11 | Claude Opus 5 (xHigh) | 86.71 | #16 (Claude Opus 5) |
| 12 | Grok 4.5 (High) | 86.55 | #54 (Grok 4.5) |
| 13 | GPT-6 Sol (Medium) | 86.52 | #19 (GPT-6 Sol) |
| 14 | MiMo-V2.6-Flash | 86.26 | #74 |
| 15 | Grok 4.7 (xHigh) | 86.23 | #26 (Grok 4.7) |
No result here: #3 Claude Opus 5.5, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=automationbench-aa-sales-raw-objective-completion · How It Works · Data refreshed daily, snapshot 2026-11-05.