Tsinghua-backed teams open-source PilotDeck 2.0, an agent workspace where switching models keeps your memory
heyshrutimishra · x · 2026-10-15
PilotDeck 2.0, an open-source agent workspace built by Tsinghua University's THUNLP, ModelBest, OpenBMB and AI9Stars, treats the model as just a setting: it connects OpenAI, Gemini, DeepSeek, Qwen, Kimi, MiniMax, and local models via Ollama (no API key), without wiping workspace memory.
- All memories are visible and editable, with permission rules controlling what the agent can access and write
- Runs locally or on a private server, keeping data and memory on user-controlled hardware
- Demoed converting images to editable Markdown, automating workflows, and building games
- Web demo, GitHub repo, and macOS/Windows clients available
More from coding & agent
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Jensen Huang details Vera Rubin agentic stack: NVL72, MGX, CMX and RTX — rohanpaul_ai · 2026-10-23
- Building an autonomous SEO agent in Hermes with three skills: programmatic pages, Keywords Everywhere, DataForSEO — EXM7777 · 2026-10-23
- Building a sports betting AI agent with onchain execution and human approval by default: architecture and lessons — Dry-Comfortable-2678 · 2026-10-23
- Capy 0.5.0 lets Claude Pro/Max subscribers reuse their plan quota, adds OIDC cloud access — garrytan · 2026-10-23
- You.com CEO Richard Socher: legacy code is human slop, let AI clean it up — RichardSocher · 2026-10-23