Running 62 Don't Train the Model, Evolve the Harness 🌿 62 Evolving an agent's harness, not its model, on Harvey's LAB
view article Article What We Learned by Reproducing 2,200 papers from ICML abidlabs • 4 days ago • 70
Running on Zero Agents Demo of Encoder-Free VLM Trained for $100 🖼 Ask questions about images and get answers instantly
Running on Zero Agents Demo of Encoder-Free VLM Trained for $100 🖼 Ask questions about images and get answers instantly
Sleeping Agents 1 Encoder-Free VLM (DenseFusion-1M + ShareGPT4V) 🖼 1 Encoder-free VLM (Qwen3-1.7B) on DenseFusion-1M + ShareGPT4V