Free profile: a model's 400 'could not be rendered into a prompt' is re-prompted to the same model with a false 'quota is back' message #200
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Seen in run 77 (production
ead9f7e, the first run on #196's fix),openspec.propose.tasksattempt 1, 2026-10-18 ~11:02Z. The session was served only by Kilo'sreflection/beam-free: four turns, the last with 34,650 characters of reasoning and awriteof the whole tasks.md. The next request got:Opencode ended the turn with
provider.invalid-request. Within 2-3 s Braid then sent the same session:It did this three times, each refused the same way within a second. The attempt then ended with tasks.md not done (
tasks: ready — 0 non-empty path(s)), and mediation started attempt 2 in a fresh session.What is wrong
retry_on_statuses, and the phrase is not inFAILOVER_ERRORS, so GoModel did not try the next model of the chain.To do
invalid_request_errorwith "chat template", "prompt rendering") toFAILOVER_ERRORS, so the router fails over at once.provider.invalid-requestfrom one model is a refusal of this session by that model. Do not park as dry. Move the session to the next model that can serve (a lane move, recorded, with its own prompt), and keep the model out of this session's chain for the rest of the attempt.Seen on production after this fix (
c1228b4, run 78,openspec.apply[6], ~14:21-14:24Z). The new behaviour works: a refusal moves the session, recorded asosf.free.movedwith its reason. One choice is poor, though:glm-5.3-flash-freerefused 3 times for capacity, so it was stalled (cause: busy, 30 min) and the step moved back to the router.zen/glm-5.3-flash-free, the model stalled as busy 90 s earlier.zen_busy.The router was dry at that moment (Gemini's minute spent, Kilo refusing, OVH's 2 a minute used), so Zen was the only target, and behind is not out. Still, a model stalled as busy minutes ago is a worse bet than waiting for the earliest reset (OVH's minute). The refusal move and
zen_forshould skip a Zen model stalled asbusyand wait, unless the wait is longer than its cool-down.Shipped in
c1228b4(changefree-refused-session-moves-on). Seen on production in run 78: Kilo's 400 moved the session to OVH (osf.free.moved, with the refusal as its why), and no quota re-prompt loop. Open: the move once chose a Zen model stalled as busy 90 s earlier (see the comment above). Design call left for the operator:free_refusedis not mediated.Left open: the OpenSpec change is not archived yet, and its tasks are not ticked against production. Deferred when the session wrapped up (2026-10-18).
Closing: verified on production (run 78,
c1228b4): Kilo's 400 moved the session to OVH with the refusal as its reason, and no re-prompt loop. The busy-Zen choice noted above is #204. The OpenSpec changefree-refused-session-moves-onarchives afterfree-profile(#164).Archived in OpenSpec as
free-refused-session-moves-on(5a953a0).