Charts / Flybridge Capital Partners
Model Agent Group Configurations and Runs
The table details various AI agent model configurations across different group tiers—such as Budget, Flagship, and Champion—and the corresponding number of evaluation runs.
| Label | Value shown on the chart |
|---|---|
| Budget - Claude Haiku 4.5 (9) | 6 |
| Budget - Gemini 3.5 Flash Lite (9) | 6 |
| Budget - GPT-5.6 Luna (9) | 6 |
| Budget - Grok 4.3 (9) | 3 |
| Budget - Qwen 3.7 Plus (9) | 3 |
| Budget - DeepSeek V4 Flash (9) | 3 |
| Flagship - Claude Opus 5 (9) | 3 |
| Champion - Claude | 9 |
| Champion - Gemini | 9 |
| Champion - GPT-5.6 | 9 |
| Final Run | 1 |
| Total | 81 + 1 Final Run |
Shown as a preview with credit and a link to the original. Publishers can request removal.
View the original on flybridge.com
