Open ErikBjare opened 3 weeks ago
I've been working on reading them with the browser tool here: https://github.com/ErikBjare/gptme/pull/196
Anthropic just announced PDF support: https://x.com/alexalbert__/status/1852394000101323193
@simonw added it to llm here: https://x.com/simonw/status/1852423593084498409
The implementation I had in mind is model-agnostic, but doesn't include the vision component (probably good enough in 80% of cases), but Anthropics solution seems to be fully multimodal.
I've been working on reading them with the browser tool here: https://github.com/ErikBjare/gptme/pull/196
Anthropic just announced PDF support: https://x.com/alexalbert__/status/1852394000101323193
@simonw added it to llm here: https://x.com/simonw/status/1852423593084498409
The implementation I had in mind is model-agnostic, but doesn't include the vision component (probably good enough in 80% of cases), but Anthropics solution seems to be fully multimodal.