| agents-on-automationbench-private-v1-0-6 | Gemini 4 Argon (Zapier AutomationBench) | #1 | Success Rate: 51.3 |
| agents-on-posttrainbench-v1-1 | Gemini 4 Argon (OpenCode) | #1 | Accuracy: 45.3 |
| chart-question-answering-on-chartography-no-tools | Gemini 4 Argon (no tools) | #1 | Pass@1: 0.716 |
| code-generation-on-deepswe | Gemini 4 Argon (mini-swe-agent) | #1 | Pass@1: 0.779 |
| video-question-answering-on-lvbench | Gemini 4 Argon | #1 | Accuracy: 91.7 |
| computer-use-agents-on-osworld-2-0-0808-offline-82-tasks | Gemini 4 Argon | #2 | Partial: 69.2 |
| code-generation-on-terminal-bench-4-0 | Gemini 4 Argon | #4 | Accuracy: 57.4 |