feature: prime train routes full_finetune TOMLs to hosted endpoint#592

Open

JannikSt wants to merge 14 commits into

mainfrom

feature/train-hosted-full-finetune

Commits on Jun 1, 2026

feature: prime train discriminates full_finetune configs and routes to hosted endpoint
JannikSt
committed
improvement: prime_cluster_id is optional in full_finetune configs
JannikSt
committed
fix: don't print per-run token on stdout — chart wires it via secretKeyRef
JannikSt
committed
fix: address PR review on full-FT dispatch
JannikSt
committed
improvement: ship full prime-rl TOML to backend instead of flat fields
JannikSt
committed
fix: move inference reasoning-content streaming changes off this PR
JannikSt
committed
fix: try-hosted-first delete avoids LoRA fallback on validation failure
JannikSt
committed
fix: typed NotFoundError replaces fragile 'HTTP 404' substring match
JannikSt
committed
fix: --output json no longer skips confirmation prompt
JannikSt
committed
Link dedicated full-FT runs to the active team
JannikSt
committed
prime train logs: expose -c trainer / -c inference / -c env-server
JannikSt
committed

Commits on Jun 2, 2026