fix: restore accuracy-first route admission

modelhub_submmit_api/README.md

@@ -79,21 +79,25 @@ bash run_poll.sh --dry-run
 ## Behavior
 
 - The runner auto-discovers all safe GPU/template combinations from the public submit catalog.
-- Automatic GPU selection uses exact 70/30 accepted-task scheduling: long-term
-  Wilson-ranked top 3 GPUs and the top GPUs from the latest 1,000 terminal tasks.
-  There is no all-GPU exploration category.
-- Within each category, weighted-fair scheduling uses estimated queue backlog hours,
+- Automatic GPU selection uses one success-first score built from qualified
+  local cohorts and ModelHub public evidence. There is no random or fixed-quota
+  exploration category.
+- Within the qualified success band, scheduling uses estimated queue backlog hours,
   recent public throughput/success, machine availability, and worker concurrency.
   Unavailable or stalled GPU pools are circuit-broken instead of continuing to absorb work.
 - Compatible frameworks are ranked by ModelHub public aggregate success statistics
   plus capped local GPU+framework evidence, with a 300-sample public minimum and
-  Wilson confidence bounds. Missing or undersized public evidence receives zero
-  traffic rather than falling back to exploration.
+  a 20% Wilson lower-bound floor. GPU pools normally require a 15% floor.
+  Missing or undersized public evidence receives zero traffic rather than
+  falling back to exploration; a low-sample GPU requires a recent success and a
+  framework that already passes the stricter framework gate.
 - New frameworks are discovered from the live catalog but get no novelty bonus.
   They are eligible only with a complete official build config that passes local
   validation and a confidence score at least 10% above the best incumbent.
-- Five consecutive local failures pause a GPU/framework pair for 12 hours; a
-  sub-20% rate over the latest 20 terminal tasks pauses it for 6 hours.
+- Five consecutive local failures pause a GPU/framework pair for 12 hours. Five
+  or more recent terminal outcomes below 10% success also pause it for 12 hours,
+  including ambiguous runtime failures. A sub-20% attributable rate over the
+  latest 20 terminal tasks pauses it for 6 hours.
 - An explicit "framework does not support this model/architecture" failure learns
   a 30-day GPU + framework + task + architecture block. Architecture identity
   comes from candidate `config.json` plus exact unsupported `model_type` or
@@ -114,20 +118,22 @@ bash run_poll.sh --dry-run
   duplicates do not advance it. The next cycle refreshes platform history before submitting again.
 - Strategy state is stored in `.modelhub_state/gpu_strategy.json`; a generation never recalculates
   during candidate submission.
-- Candidate discovery starts with the configured recent window, then automatically expands to
-  7 days, 30 days, and the first 3,000 raw ModelScope rows ordered by update time. If the
-  candidate pool is still short, it rotates through up to 200 download-ranked rows per cycle
-  as an older-model fallback.
+- Candidate discovery starts with the configured recent window, then
+  automatically expands to 7 days, 30 days, and the first 3,000 raw ModelScope
+  rows ordered by update time. Download popularity is not treated as ModelHub
+  compatibility evidence.
 - Model verification responses are cached across poll cycles for 15 minutes. Local model/GPU
   failures cool down after 24 hours instead of remaining permanently blocked.
-- Community deduplication is model/GPU-specific: another GPU's adaptation does not block the
-  current GPU. `未验证` with no records is eligible. `验证失败` is eligible only with a
-  different publicly vetted framework after all recorded attempts explicitly failed;
-  `已验证`, `验证中`, and ambiguous records remain blocked. Every actual submission
-  performs a fresh uncached check for its exact GPU and framework.
+- Community deduplication is model/GPU-specific: another GPU's adaptation does
+  not block the current GPU. Only `未验证` with no records is eligible;
+  `验证失败`, `已验证`, `验证中`, and ambiguous records remain blocked. Every actual
+  submission performs a fresh uncached check for its exact GPU.
 - If the community lookup is unavailable, submission is deferred. A platform model-uniqueness
   rejection permanently excludes only that model/GPU combination from future local retries.
-- The same framework is not retried for a model/GPU after a recorded failure.
+- A model may use at most six GPU routes across poll cycles, and at most two
+  locally submitted routes for that model may await feedback at once. The first
+  six ranked GPUs remain preferred; lower-ranked eligible GPUs are inspected in
+  order only when that tranche yields no safe candidate.
 - Multiple ModelHub tokens are pooled and used to route submissions to the account with available async capacity.
 - Concurrent submissions reserve account slots locally, and an account-capacity race automatically falls through to another account.
 - Concurrent local processes claim model/GPU pairs in `.modelhub_state/submission_claims.jsonl`; shared ledger, history, and outcome files use process locks and atomic replacement.