zeekay 06f8e7b472 zen-router: real local-usage dataset, expanded catalog, measured results ...

- training/catalog.yaml: routable universe (24 models) = 3 local zen + zen cloud
  tiers + Anthropic/OpenAI/Gemini/Grok/DeepSeek/Kimi, priced from the gateway
  live config (hanzoai/ai/conf/models.yaml). Route head auto-sizes 7 -> 24.
- training/extract_local_sessions.py: build eval rows from ~/.claude + ~/.codex
  session logs; real model ids mapped 1:1 to catalog ids (no generic tiers).
- sft.py: configurable dtype (fp32 stable on MPS), record backbone base in
  router_config.json.
- eval_routing.py: device-aware, single-forward latency + pred distributions.
- config.local.yaml: reduced MPS run off on-disk zen-nano-0.6b.
- docs/config.md: drop-in router_policy.yaml (zero-code cost routing) +
  route-to-anything (one catalog row).
- README Measured: M4 Max, 5455 rows (94.6% opus imbalance), task_acc 0.824 /
  route_acc 0.961 / top3 0.999, single-forward 137ms, with full caveats.

2026-07-07 17:12:45 -07:00

..

config.md

zen-router: real local-usage dataset, expanded catalog, measured results

2026-07-07 17:12:45 -07:00

integration.md

zen-router: tiny trainable routing model

2026-07-07 15:13:57 -07:00