docs.dev vs Needle2

Point your coding agent at your repo and your docs; it opens a docs branch, your team reviews on the rendered page, publishing is a commit. Runs on your Cloudflare, no...

An open 45M-parameter model for tool calling, device use, and structured extraction. Needle 2 runs as a 14 MB binary in 28 MB of session RAM.

About docs.dev

Point your coding agent at your repo and your docs; it opens a docs branch, your team reviews on the rendered page, publishing is a commit. Runs on your Cloudflare, no platform fee, no metered AI. Reviews, branches, and rollbacks come free, because it is just git. Your content in their database, and an export button you hope works when you leave. Writing AI is the agent you already have; reader-facing AI runs on your Workers AI.

About Needle2

An open 45M-parameter model for tool calling, device use, and structured extraction. Needle 2 runs as a 14 MB binary in 28 MB of session RAM. Needle 2 is measured end-to-end through the shipped binary at CQ2-bit deployment precision with tool retrieval on; baselines run the released checkpoints under vLLM, and Apple FM runs on-device. Bringing On-Device AI to <$200 Devices : Edge AI has lately meant Macs and PCs, but the true edge is mostly cheap hardware: there are more than 21 billion IoT devices against roughly 1.5 billion PCs.