| ▲ | GPT-5.6 vs. Claude Fable 5 for Physical AI, which performs best?(juliahub.com) | |
| 14 points by mbauman an hour ago | 3 comments | ||
| ▲ | grim_io 7 minutes ago | parent | next [-] | |
I'd expect google to do well here, since they were historically strong at multimodal and physics. | ||
| ▲ | jespinel 7 minutes ago | parent | prev | next [-] | |
Nice! Is is missing Codex in the agent harnesses comparison IMO. | ||
| ▲ | gizmodo59 4 minutes ago | parent | prev [-] | |
Yet another "benchmark to promote their own harness" | ||