CoolBoy/submmit
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

feat: learn framework architecture incompatibilities

Browse Source
This commit is contained in:
CoolBoy
2026-08-12 08:02:51 +08:00
parent d908706f9a
commit 615bcad124
10 changed files with 503 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
README.md
View File

@@ -110,6 +110,19 @@ idle card does not permanently poison otherwise successful evidence.
Candidate shortages expand the model search window; they never unlock an
unvetted GPU or framework.
Failed-task archives are also classified conservatively. When ModelHub explicitly
says that the selected framework does not support the model or architecture, the
runner learns an exact GPU + framework + task type + architecture block from the
candidate repository's `config.json`. Repository names are never used as
architecture evidence. Exact `architectures` values take priority and
`model_type` is used only when `architectures` is absent; missing metadata does
not create a block. Generic unsupported operators, attention backends, GPU types,
quantization failures, and OOMs cannot enter this blacklist. A newer success for
the same exact combination clears the block, and otherwise it expires after 30
days. Set `MODELHUB_ARCHITECTURE_BLOCK_TTL_DAYS` to a value from 1 to 365 to
change that window. The stats report exposes `architectureCompatibilityBlocks`
and per-GPU/framework block counts.
Before a candidate reaches the submit queue, failure-informed preflight checks
the actual ModelScope repository structure and file sizes. Non-GGUF text
frameworks require root-level config, weights, and tokenizer assets. The memory
@@ -253,12 +266,16 @@ and startup cleanup, then limit minus 5 for scheduled dynamic cleanup.
Version `2026.08.12.1` protects running tasks from age-based cleanup and records
worker-initiated stops as `policy_cancelled`, excluding them from success-rate,
failure-cooldown, and circuit-breaker evidence.
Version `2026.08.12.2` learns conservative, expiring GPU/framework/architecture
compatibility blocks only from explicit ModelHub failure text, matches candidate
`config.json` metadata instead of repository names, and lets newer success
evidence clear stale blocks.
## Deploy
Create a tag and submit the repository URL plus tag in "我的适配智能体".
```bash
git tag agent-v19
git push origin agent-v19
git tag agent-v20
git push origin agent-v20
```