mirror of
https://github.com/kvcache-ai/ktransformers.git
synced 2026-10-27 12:28:19 +08:00
yyj
7c35863f62
* ci: add isolated model E2E acceptance for PR wheels and PyPI * ci: recheck GPU occupancy before each model starts * ci: disable inherited container healthchecks during acceptance * ci: wait for free GPUs before native import validation * docs: record qj5090 acceptance infrastructure validation * ci: add immutable candidate and public PyPI release acceptance gates
25 lines
693 B
JSON
25 lines
693 B
JSON
{
|
|
"schema": 1,
|
|
"image": "registry.example/kt-model-e2e@sha256:REPLACE_WITH_REVIEWED_IMAGE_DIGEST",
|
|
"lock_path": "/var/lock/ktransformers/qj5090-model-e2e.lock",
|
|
"queue_timeout_seconds": 21600,
|
|
"test_timeout_seconds": 43200,
|
|
"models": {
|
|
"qwen3": {
|
|
"path": "/mnt/data/models/Qwen3-30B-A3B-Instruct-2507",
|
|
"revision": "REPLACE",
|
|
"config_sha256": "REPLACE"
|
|
},
|
|
"deepseek_v31": {
|
|
"path": "/mnt/data/models/DeepSeek-V3.1",
|
|
"revision": "REPLACE",
|
|
"config_sha256": "REPLACE"
|
|
},
|
|
"glm53": {
|
|
"path": "/mnt/sft_yyj_yyj/models/GLM-5.3-Flash-public-04c4e9e9",
|
|
"revision": "REPLACE",
|
|
"config_sha256": "REPLACE"
|
|
}
|
|
}
|
|
}
|