hmm-model-v1
weave-os/routerhmm-model-v1Jul 13, 2026by steventohme
Release notes summary
This release introduces a portable, frozen HMM router artifact designed for the companion policy sidecar. Built with the google/gemini-embedding-2 model (3072 dimensions), it is a runtime-only artifact optimized for the self-hostable HMM sidecar PR.
Key Highlights
- Portable frozen artifact for self-hostable HMM sidecar.
- Uses google/gemini-embedding-2 model (3072 dimensions).
- Runtime-only with no training data or registry dependencies.
New Features
- Portable frozen HMM router artifact for the policy sidecar.
- Support for google/gemini-embedding-2 embedding contract.
- Explicit configuration option to route via HMM (defaulting to cluster routing).
Full Release Notes
Portable frozen HMM router artifact for the companion policy sidecar. - Model: `hmm-router-frozen-v1` - Embedding contract: `google/gemini-embedding-2`, 3072 dimensions - SHA-256: `b326bda644aac86831bae47d0b69de008046f62324a4b93a1eba702581c174df` - Runtime data only: no Python pickle, training rows, prompts, training provenance, or WorkWeave registry dependency This prerelease is intended for the self-hostable HMM sidecar PR. The router remains on cluster routing unless HMM is explicitly configured.
Related Repositories
Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
A fast reverse proxy to help you expose a local server behind a NAT or firewall to the internet.
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
The world’s fastest framework for building websites.
:cherry_blossom: A command-line fuzzy finder