Jev directory
CommunityJev Repos & SDKs
Jev Model Calling Project
Chinese-language GitHub project that wraps Jev's /predict endpoint behind a CLOUDSOA-HMAC-SHA256 gateway signature. Ships jev_predict_common.py for signed requests and jev_client.py, a CLI with weather and web rerank examples.
- Stars
- 0
- Reading time
- 2 min
- Published by
- Community
- Added
- Oct 2026
Highlights
- jev_predict_common.py wraps the /predict endpoint with a CLOUDSOA-HMAC-SHA256 gateway signature.
- jev_client.py is a command-line script with built-in weather and web rerank examples and custom payload support.
- The payload data field matches Jev's official systemone request body, so the same questions work against the cloud API.
- Documents noul, choice, and score primitives, confidence tiers, and threshold design in Chinese.
- Repository has 0 stars and is maintained by the individual account dulin20240919-dot.
Questions about this entry
Short answers drawn from the linked project or page.
Two Python files and a README: jev_predict_common.py, a signed /predict request module, and jev_client.py, a command-line script with built-in weather and web rerank examples that also accepts custom payloads.
All three primitives. The weather example sends noul, choice, and score questions in one request; the rerank example uses a score question to rate how well a document answers a query on a 0-9 scale.
The payload's data field is identical to Jev's official systemone request body, and the project adds a CLOUDSOA-HMAC-SHA256 signature layer around it. The digest does not document how signing credentials are supplied.
The scripts are Python and the README and reference notes are written in Chinese, compiled from an internal usage document and the official Jev documentation.
No. It is a personal repository by dulin20240919-dot that documents and wraps calls to the Jev model, assembled from an internal usage document and the official TypeSafe documentation.
It puts a query plus document fields such as title, chunk, site, and publish time into state, then asks a score question for how helpful the document is for the query, producing 0-9 relevance scores for reordering results.
Jev accepts text only, and its primary training language is English, so Chinese inputs work but with lower accuracy. The notes suggest writing instructions and criteria in English and setting conservative thresholds.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
Facebook: Intake Router keeps diagnosis out of the answer space at the front desk.
FacebookClinicnotes
X: OpenRouter Decision Model Rankings
X: herdr + jev, an agent router
X: A group chat bot with decision room routing
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.