gemma4-research/README.md

gemma4-research

Research corpus and implementation guidance for Google Gemma 4, based on production use in Seth's homelab.

Files

File	What	When to Read
SYNTHESIS.md	Start here. Opinionated guide — how to build with Gemma 4	Before any new Gemma 4 implementation
GOTCHAS.md	Known issues and workarounds, severity-ranked	When debugging Gemma 4 issues or starting a new project
IMPLEMENTATIONS.md	Patterns from Simon and AI_Visualizer	When designing a new Gemma 4 integration
CORPUS_architecture.md	Model architecture details (layers, attention, PLE, MoE)	When you need to understand WHY Gemma 4 behaves a certain way
CORPUS_ollama_variants.md	Available models, sizes, VRAM, Ollama settings	When choosing a model variant or configuring Ollama
CORPUS_capabilities.md	Modalities (vision, audio, video, tools), what it can/can't do	When scoping what Gemma 4 can handle
CORPUS_benchmarks.md	Full benchmark table vs Gemma 3, arena scores, agentic scores	When comparing Gemma 4 to alternatives
CORPUS_tool_calling_format.md	Native token format + JSON API format for function calling	When implementing tool calling
CORPUS_tool_calling_2026-05.md	May 2026 update — read first for tool-call debugging. Ollama parser fixes (#15539), think flag properly characterized, the "doesn't fire tools without explicit ask" research (Probe & Prefill, BiasBusters, When2Call), XGrammar-2. Supersedes parts of GOTCHAS.md (think:false rule, Vulkan loop, streaming bug all updated in-place with pointers here)	When debugging tool-call reliability, deciding think value, or planning a router/probe layer
CORPUS_agentic_retrieval.md	Post-RAG landscape — hybrid retrieval + cross-encoder rerank as the settled base, CRAG-shaped flows in LangGraph/LlamaIndex, GraphRAG (LightRAG/LazyGraphRAG), memory architectures (mem0 vs Letta MemFS), deep-research agents (gpt-researcher, smolagents), production-vs-experimental sorting	When stepping up from basic embedding RAG to agentic retrieval, picking a memory layer, or scoping a deep-research agent
CORPUS_cli_coding_agent.md	Positioning Gemma 4 for CLI coding agent use (openclaw / open code / pi / hermes / aider style). Honest take on what Google did and didn't measure, head-to-head with qwen3-coder:30b, homelab setup pointer	When scoping a CLI coding agent or deciding Gemma 4 vs Qwen3-Coder
docs/openwebui-setup.md	How to configure Gemma 4 inside OpenWebUI — per-setting reference, two ready-to-bake Workspace Model profiles (chat + extract), and a symptom→cause troubleshooting table mapped back to GOTCHAS.md. Assumes Ollama + OpenWebUI are already running.	When setting up or debugging a Gemma 4 model in OpenWebUI, or handing the front-end config to someone else
docs/reference/bakeoff-2026-04-18.md	CLI-coding-agent bakeoff on 3090 Ti. Rounds 1/2 misidentified the cause; Round 3 (the correct one): think: false silent-stops gemma4:26b at certain multi-turn states on 32K context. 31B and Qwen3-Coder robust to the flag. Harness at scripts/bakeoff/	When deciding which model to back a CLI agent with, writing a custom agent payload, or debugging a silent tool-call halt
docs/reference/mort-bakeoff-2026-04-18.md	mort-bot-specific think=true vs think=false bakeoff on mort's actual loop shape (gemma4:26b, num_ctx=8192). Thinking does NOT accumulate in context on Ollama 0.20.4 — strips it from serialized history. Both settings behave identically on step counts, tool counts, wall clock. Harness at scripts/mort-bakeoff/	When deciding mort-bot's THINK env var, or when someone claims "think=true eats context" without pinning an Ollama version
docs/reference/diffusiongemma-smoketest-2026-06-17.md	DiffusionGemma (google/diffusiongemma-26B-A4B-it, released 2026-06-10) — Google's first open-weight text-diffusion LLM (26B-A4B MoE + canvas diffusion head). Research + first-hand smoke test on steel141's 3090 Ti: does NOT run in Ollama (unknown model architecture: 'diffusion-gemma'), needs llama-diffusion-cli from ggml-org/llama.cpp PR #24423. Build recipe, throughput (~106 tok/s effective / ~2030 tok/s in-step-parallel at Q4_K_M), the nvidia-smi-vs-CUDA device-ordering gotcha, and the "thought channel eats the canvas" behavior.	When evaluating a text-diffusion model, deciding if DiffusionGemma is homelab-deployable, or reproducing the llama-diffusion-cli build
docs/reference/gpu-bakeoff-2026-04-20.md	Cross-GPU throughput bakeoff: steel141 RTX 3090 Ti vs strix-halo (AMD Strix Halo). 3090 Ti wins decode decisively (128 tok/s on 26B MoE). Strix gets ~42% of that on ~25% of the bandwidth. Also quantifies the MoE vs dense gap: 26B decodes ~4.7× faster than 31B on both cards. Harness at scripts/gpu-bakeoff/	When choosing which host to run a Gemma 4 workload on
tooling/	Canonical upstream tooling — real scripts, notebooks, model cards, and configs pulled from Google / HF / framework maintainers (147 files). Subdirs: google-official/, huggingface/, inference-frameworks/, gemma-family/, fine-tuning/. See tooling/README.md for index and findings that update the older CORPUS_* docs	When you need authoritative source material — model cards, chat templates, fine-tuning recipes, serving commands for vLLM / llama.cpp / MLX, or to scope a specialized sibling (ShieldGemma, EmbeddingGemma, etc.)

Source Projects

• Simon (~/bin/FreibergFamily/simon/) — Multi-turn chat agent with 6 tools, genealogy historian
• AI Visualizer (~/bin/AI_Visualizer/) — Music video generator, 4-stage Gemma pipeline + vision

Key Insight

Gemma 4 is ultra-compliant and highly capable but doesn't know who it is. It needs explicit system prompts, not hand-holding. Due to fast local inference, sequential tool calls beat long JSON requests.