TLDR AI · RESEARCH & INNOVATION
LLAVA-O1: REASONING VLMS
Summary from the TLDR AI newsletter
While this is not an actual replication of the process used to train o1, it is still a strong VLM trained on reasoning traces.
TLDR AI · RESEARCH & INNOVATION
While this is not an actual replication of the process used to train o1, it is still a strong VLM trained on reasoning traces.