docs: refresh Gemma 4/Gecko → Gemma 3n + EmbeddingGemma - #70
Merged
Conversation
The user-facing docs and staging scripts still described the pre-upgrade stack. Brings them in line with what main actually ships (Gemma 3n generator + EmbeddingGemma-300M retriever, RAG bundle v0.3.0). - README.md: overview, architecture diagram, retrieval flow, model-download table (self-hosted ungated mirrors + pinned SHA-256), sync script comments, and the eval/latency section (Gemma 3n now the deployed generator; EmbeddingGemma retrieval latency). - CLAUDE.md: embeddings component, model-files list, and embeddings backend note (EmbeddingGemma is CPU/XNNPACK only via the custom embedder; Gecko retained as a config fallback). - device_push/README.md: model table + bundle counts (87 PDFs / 63,650 chunks), current vs legacy models. - scripts/sync_models.sh + push_to_device.sh: fetch/push the EmbeddingGemma model + tokenizer instead of Gecko; --gecko-only → --embedder-only. Left untouched intentionally: evaluation/* (the harness legitimately lists Gemma 4 as a comparison model + historical runbooks), and app-code comments that correctly describe Gecko as the legacy fallback path. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Updates the user-facing docs + staging scripts that still described the pre-upgrade stack (Gemma 4 + Gecko) to match what
mainships: Gemma 3n generator + EmbeddingGemma-300M retriever, RAG bundle v0.3.0.README.md,CLAUDE.md,device_push/README.md: architecture, model tables, download sources (ungated mirrors + pinned SHA-256), counts (87 PDFs / 63,650 chunks), eval/latency notes.scripts/sync_models.sh+push_to_device.sh: stage/push EmbeddingGemma model + tokenizer;--gecko-only→--embedder-only.evaluation/*(legit comparison-model + historical runbook refs) and app-code comments describing Gecko as the legacy fallback.🤖 Generated with Claude Code