Changelog

All notable changes to Cerebro are documented here. Format follows Keep a Changelog. Versioning follows PEP 440.

---

[1.0.0b1] - 2026-04-05

Added

Anthropic Claude provider (CEREBRO_LLM_PROVIDER=anthropic or claude)

• Default model: claude-sonnet-4-6
• Embeddings delegated to local EmbeddingSystem (Anthropic has no embeddings API)
• Config: ANTHROPIC_API_KEY, ANTHROPIC_MODEL, ANTHROPIC_TIMEOUT
• Install: poetry install --with anthropic

Groq provider (CEREBRO_LLM_PROVIDER=groq)

• Default model: llama-3.3-70b-versatile
• Embeddings delegated to local EmbeddingSystem (Groq has no embeddings API)
• Config: GROQ_API_KEY, GROQ_MODEL, GROQ_TIMEOUT
• Install: poetry install --with groq

Gemini provider (CEREBRO_LLM_PROVIDER=gemini or google-gemini)

• Default generation model: gemini-1.5-flash
• Native embeddings via google-generativeai SDK (models/text-embedding-004)
• Config: GEMINI_API_KEY, GEMINI_MODEL, GEMINI_EMBEDDING_MODEL, GEMINI_TIMEOUT
• Install: poetry install --with gemini

Nix packaging

• packages.default in flake.nix via poetry2nix.mkPoetryApplication — enables nix build and nix run
• nix/cerebro.nix — standalone buildPythonPackage derivation for nixpkgs PR submission

New provider aliases registered in RigorousRAGEngine:

Alias	Canonical
anthropic	anthropic
claude	anthropic
groq	groq
gemini	gemini
google-gemini	gemini

Changed

• Version bumped from 0.1.0 to 1.0.0b1 (BETA launch)
• src/cerebro/__init__.__version__ aligned to 1.0.0b1 (was inconsistently 2.0.0)
• .env.example documents env vars for all three new providers
• Justfile gains a release recipe and a todo progress tracker

---

[0.1.0] - Initial release

• Local-first RAG engine with llama.cpp + ChromaDB
• OpenAI-compatible provider adapter
• Optional Google Vertex AI integration
• React dashboard + Textual TUI
• GCP credits monitoring (optional)