OpenBMB open-sources MiniCPM5-2B, top agentic model under 4B that runs offline
thetripathi58 · x · 2026-09-12
OpenBMB has open-sourced MiniCPM5-2B, a dense 2B-parameter model built for reasoning, coding, and tool use on resource-constrained hardware — capable of running agents fully offline on a laptop or even a phone.
- It tops Artificial Analysis's Agentic Index among models under 4B with a score of 20, vs 9 for Granite 4.2 8B
- Beyond weights, OpenBMB is also releasing training methods, agent data, and parts of the RL framework
- Blogger akshaypachaar tested it locally on a constrained CI repair agent, tasking it to debug an idempotency-key retry bug where totals jumped from $109 to $118
- It is notably strong at coding and tool calling, showing how capable small models have become
Related event: OpenBMB's MiniCPM5-2B Tops Sub-4B Open Models in Intelligence Index(4 posts)→
More from coding & agent
- BELL: Open-Source Stack Combines Browserbase, Eve, iMessage and Stripe Link for Personal Agents — jeff_weinstein · 2026-10-27
- Legora runs 2,000 parallel agent evals on Modal Sandboxes for legal AI — blelbach · 2026-10-27
- Two engineers rebuilt a FedEx supplier's billing core in 3.5 months vs an estimated 7-8 — alex_verem · 2026-10-27
- GPT-6.1 Terra scores 48 on AA Coding Agent Index at one-third the cost of GPT-5.6 Terra — ArtificialAnlys · 2026-10-27
- Shift-1 router cuts coding agent spend up to 55% while keeping SWE-bench accuracy — ypatil125 · 2026-10-27
- Non-coder vibe-codes a Godot JRPG with Claude Code in two weeks, music and art synced — ADHDavid · 2026-10-27