Repository Issue Activity (beta)

#432 MTP docs: draft/target roles reversed in the pending-tokens caption

#415 the minimum version of transformers required to support Gemma 4?

#273 FunctionGemma: Fine-tuned model outputs different format than base model

#399 Gemma 4 Concurrent on DGX Spark cluster

#380 [Feature Request]: Reliable local Gemma assistants with structured output, source boundaries, and no-write evaluation

#346 [Gemma 4] `use_cache=False` corrupts attention computation, producing garbage logits

#372 Gemma 4 breaks when trying to search with Ollama Web Search tool

#225 Seems wrong answer in [Gemma_2]Finetune_with_Function_Calling.ipynb

#263 Finetuning of gemma2 model using LORA

#256 Less than 1.91b parameters

#156 Develop Gemma 3 chatbot with Llama.cpp & Gradio

#190 Full tools implementations

#236 Ideas for GEMMA 4

#307 Request: Tool calling for Gemma 4 series

#309 Request: Tool calling for Gemma 4 series

#308 Request: Tool calling for Gemma 4 series

#291 [Feature Request]: Speculative Decoding with Gemma 3 for 2-3x Inference Speedup

#161 Only instruct_finutune achieve domain knowledge injection? It upended my perception...

#275 [Feature Request]: Modernize JAX/TPU Parallelism with Gemma 3 (Kaggle v5e-8 Support)

#270 [Proposal] New Notebook: Zero-Shot Visual Document Extraction and JSON Parsing with Gemma 3

#266 Feature Request: Smart Contract Auditing & Secure Refactoring with Gemma 2

#157 Gemma Model Fine-tuning UI: Streamlit-Based Real-Time Training and Custom Model Export

#154 Mechanistic Interpretability: Logit Lens & Attention Analysis for Gemma 2B Models

#158 Gemma 3 Finetuning Cookbook

#274 FunctionGemma: Conversion to .litertlm for on-device deployment - unable to install dependencies