EN FR

Layer 2 — Pool

Backend · OpenAI-compatible universal API · orchestration · zero-knowledge memory

Users / Agents Browser · OpenCode · ClawCode Cursor · aider · iamine CLI Cloudflare + nginx TLS Flexible SSL (→ Full strict TODO) :443 HTTPS · :443 WSS FastAPI — OpenAI-compatible universal API /v1/chat/completions · /v1/models · /v1/federation/* · /v1/opencode-md SSE streaming · tool-calls · auto conv_id · admin_token Auth & Tokens email/Google login email verification (M16) iam_token per user separate admin_token Cooperative routing deficit scoring GPU 61% / CPU 38% always answer (no 503) unknown_model auto-exclude LLM Checker core/checker.py periodic bench skip tool-calls DB config pool_config Sub-agents auto-review (1s) cross-pool (M7a) multi-role pipeline SSE expose review Zero-knowledge memory (4-tier, M13) L1 messages observ. TTL raw JSON hot L2 conversations Fernet encrypted compaction opt-in L3 embeddings pgvector user facts filtered RAG L4 hybrid MCP server agent memory federation sync PostgreSQL + pgvector pool brain · DB-first for all persistent data • users · accounts · tokens · sessions • messages · conversations · memory (L1/L2/L3/L4) • federation_state · peers · capabilities · gossip_log • revenue_ledger · disputes · slashing_events • pool_config (admin) · migrations 001 → 019+ WebSocket pool wss://cellule.ai/ws workers join/leave heartbeat · job dispatch 4 workers · 0 busy · 39 pending Admin dashboard admin.html + admin_pool.html DB cleanup · blacklist · PG accounts LLM config · FR tooltips · EN TODO admin_token auth cellule.ai Frontend trial chat (pinned 2B model) user dashboard · tokens · tools interactive molecule canvas community wording (no "free") WORKERS (atoms) — Qwen 14B · 3B · 9B · 30B MoE · heterogeneous · multi-machine · Trust-3 reset ← click to see layer 1 atom → Federation (layer 3) anti-entropy gossip · merkle · RAID quorum outbound to other bonded pools → Economy (layer 4) revenue_ledger · settlement · slashing · $IAMINE fed by each inference read/write dispatch job

Universal API

  • • OpenAI-compatible (chat/completions)
  • • SSE streaming tokens + tool-calls
  • • auto conv_id (L2 persistence)
  • • clients: OpenCode, Cursor, aider, ClawCode

DB-first

  • • All persistent data in PG
  • • Pool memory = cache
  • • pgvector for RAG/L3
  • • Versioned migrations 001→019+

Cooperative routing

  • • Deficit scoring (GPU/CPU fairness)
  • • "Always answer" doctrine
  • • No outbound 503
  • • Auto-exclude unknown_model

Sub-agents LIVE

  • • Auto-review (1s, from phase 1)
  • • Cross-pool forward (M7a)
  • • Multi-role pipeline (N-pool scale)
  • • SSE exposes review to clients

Zero-knowledge memory

  • • L2 conversations Fernet encrypted
  • • User opt-in (toggle)
  • • GDPR export + delete
  • • Pool admin cannot read content

LLM agents (Z2)

  • • RED Qwen3-30B-A3B (admin)
  • • Coder · Tank · Scout (dedicated)
  • • RED daemon 3h + self_update
  • • 4 active workers today