Gemini 4 Neon lands Day 0 in LangChain's Managed Deep Agents for long-running agent work
hwchase17 · x · 2026-11-19
- Google released Gemini 4 Neon, positioned for long-running, tool-heavy agent workloads, with Logan K calling it the right balance of intelligence, speed, and usability; it's live in Google AI Studio and the API.
- LangChain's Managed Deep Agents shipped Day 0 support: one line of config (model="googlegenai:gemini-4-neon", requires langchain-google-genai>=4.5.1) wires the model into the agent.
- MDA (public beta on LangSmith Cloud and BYOC) hosts the Deep Agents harness—planning, tool calls, filesystem, subagent delegation—plus managed runtime, proactive runs (schedules, memory, channels), and opt-in Harbor evals, so developers only write the agent's instructions, tools, and skills.
More from coding & agent
- Higgsfield Launches MCP 2.0 for Claude, Bringing Its Creative Studio Into Chat — piotrbinkowski · 2026-11-19
- CoRL workshop: running Karpathy's autoresearch on real robots — the sandbox, not the model, is the bottleneck — Kyrannio · 2026-11-19
- OpenAI data: 90th-percentile researcher now spends $23,000 on agentic AI tools — kliu128 · 2026-11-19
- Web search grounding cut a calling agent's booking accuracy by 5+ points across 2,415 tests — Evening_Hawk_7470 · 2026-11-19
- badlogic: agents 'reward-hack' past your lint rules and tests—without insight under the hood, your verifications may be worthless — teropa · 2026-11-19
- Color skill flips GPT-6.2 Astra's TiKZ unicorn art from cartoon to flat vector style — Angaisb_ · 2026-11-19