Instructions to use google/gemma-4-E2B-it with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use google/gemma-4-E2B-it with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("google/gemma-4-E2B-it") model = AutoModelForMultimodalLM.from_pretrained("google/gemma-4-E2B-it", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- AMD Developer Cloud
Commit History
-
Add response_template to tokenizer_config.json (#40)
- Add response_template to tokenizer_config.json (6b72efbd2328cb6b042f28b46b0d29d9394f3778) - Update README.md (b3da59c6fe893e5932073fa7aa473c4ff5cd2768) Co-authored-by: Matthew Carrigan <Rocketknight1@users.noreply.huggingface.co>3e22461
-
Update README (#42)
- Update README (ebfa5cfaf9ca4e695f227d136e847f814841961c)179516f -
fix: chat template — null handling, reasoning preservation, turn-tag balance, input validation (#35)
- fix: chat template — null handling, reasoning preservation, turn-tag balance, input validation (c3cdbdc9e7fd33008f19e1e1d8d46382b1349cac) - fix: restore model turn + thinking cue after tool responses (56ced01461213f76bf593d8d3a373c836b505615) - fix: prevent extra <turn|> when assistant has content + tool_calls + continuation (803d0d827f7d0bac388a5c9e4544838640539275) - fix: revert add_generation_prompt regression + preserve_thinking default (89ef9a8132ff3a2e2e92dc6f6a89768e9c203ddf) - fix: render thinking channel regardless of tool_calls presence (116c59c582efe06103309f65cfc3685ff21ab9f0) - fix: update stale comment — thinking channel no longer tool-call-only (b515064b63ff28985d549455f7709f112e8a5e39) - fix: scope preserve_thinking to tool-call turns only (25cea386e04b30ba321544e29109660b535401e8) - fix: scope preserve_thinking to tool-call turns, remove extra newline in continuation turns (f2a612070c459f37cd3a1ae0229376758ecf25be) - Add canonical header to chat template (35d9ac00a35…8995379
-
Add Technical Report
Verified signature9dbdf8a
-
Update README.md
Verified signature70af34e
-
Update README.md
Verified signature8dfc656
-
fix(chat_template): Emit multimodal placeholders in tool response content-parts (#28)
- fix(chat_template): Emit multimodal placeholders in tool response content-parts (ffcc5ed8beb1f25f12960cde3b1fd7c9b765794e) Co-authored-by: Harshal Janjani <harshaljanjani@users.noreply.huggingface.co>905e84b
-
Fix lineage (#31)
- Fix lineage (a5c9ec7042a932c0890300888fe89facf2e5ad88) Co-authored-by: Omar Sanseviero <osanseviero@users.noreply.huggingface.co>b324173
-
fix(chat_template): update SI and tool call handling (#26)
- fix(chat_template): update SI and tool call handling (291f4c095bb92d9293c405100babf9a6e5afaa52) Co-authored-by: Douglas Reid <dougreid@users.noreply.huggingface.co>6b7e72c
-
Update response schema (#15)
- Update response schema (7c7816477f0892c3ae92f73c0c59e5b9dda1689e) Co-authored-by: Matthew Carrigan <Rocketknight1@users.noreply.huggingface.co>b4a6011 -
fix: prefix-preserving tool calls + dialog compliance (#13)
- fix: prefix-preserving tool calls + dialog compliance (3e6d8487990c10212a4238741214fc128e6f720f) Co-authored-by: Luciano Martins <lucianommartins@users.noreply.huggingface.co>b446025
-
Preparing for release!
Co-authored-by: Rocketknight1 <Rocketknight1@users.noreply.huggingface.co> Co-authored-by: RyanMullins <RyanMullins@users.noreply.huggingface.co> Co-authored-by: pcuenq <pcuenq@users.noreply.huggingface.co> Co-authored-by: bebechien <bebechien@users.noreply.huggingface.co> Co-authored-by: osanseviero <osanseviero@users.noreply.huggingface.co> Co-authored-by: douglas reid <douglas reid@users.noreply.huggingface.co>4742fe8