Add Gemma license gate and publish local model assets

2026-10-19 12:02:45 +02:00
parent 440652214c
commit 73c0c0b8ba
16 changed files with 699 additions and 7 deletions
@@ -0,0 +1,21 @@
+Model: google/gemma-3-1b-it
+Distributed GGUF repository: ggml-org/gemma-3-1b-it-GGUF
+Distributed file: gemma-3-1b-it-Q4_K_M.gguf
+Repository revision: f9c28bcd85737ffc5aef028638d3341d49869c27
+Pinned upstream URL: https://huggingface.co/ggml-org/gemma-3-1b-it-GGUF/resolve/f9c28bcd85737ffc5aef028638d3341d49869c27/gemma-3-1b-it-Q4_K_M.gguf
+File SHA-256: 8ccc5cd1f1b3602548715ae25a66ed73fd5dc68a210412eea643eb20eb75a135
+
+The upstream GGUF repository provides a quantized conversion of Google's
+Gemma 3 1B Instruct model. Model use and redistribution remain subject to the
+Gemma Terms of Use and Gemma Prohibited Use Policy distributed with this file.
+
+Runtime components:
+- Wllama 3.8.1, npm package @wllama/wllama; MIT; see WLLAMA_LICENSE.txt.
+- llama.cpp WebAssembly backend; MIT; see LLAMA_CPP_LICENSE.txt.
+
+Legal text snapshots checked against Google's official pages on 2026-10-19:
+- https://ai.google.dev/gemma/terms/ (page states last modified April 1, 2026)
+- https://ai.google.dev/gemma/prohibited_use_policy (page states last modified February 21, 2024)
+The bundled HTML contains the full article text in a minimal, script-free wrapper.
+The manifest fingerprints the exact bundled copies; a changed copy requires new
+user acceptance before this browser downloads or uses Gemma.