by KaLM-Embedding · Agent Tool · ★ 37
KaLM-Jev A local Choice / Score / Noul service built on KaLM-Reranker-V1-R2. It uses PyTorch and Transformers, runs one model per process, and returns structured judgments without generating answer text. The underlying reranker is described in KaLM-Reranker-V1: Fast but Not Late Interaction for Compressed Document Reranking. Demo: Try KaLM-Jev in the Hugging Face Space. Models: Explore the available checkpoints in the Lychee KaLM Reranker collection. Installation and quick start Requires Python 3.10+. Run the following commands from the repository root.
| Stars | 37 |
| Forks | 2 |
| Language | Python |
| Category | Agent Tool |
| Quality Score | 62.7697518180499/100 |
| Last Updated | 2026-09-21 |
| Created | 2026-09-21 |
| Platforms | python |
| Est. Tokens | ~15k |
These tools work well together with KaLM-Jev for enhanced workflows:
Explore other popular agent tool tools:
KaLM-Jev is Meet KaLM-Jev — your local, Jev-style judgment engine, available in Nano, Small, and Large.. It is categorized as a Agent Tool with 37 GitHub stars.
KaLM-Jev is primarily written in Python.
You can find installation instructions and usage details in the KaLM-Jev GitHub repository at github.com/KaLM-Embedding/KaLM-Jev. The project has 37 stars and 2 forks, indicating an active community.