Living Models paired Gemma 4 with BOTANIC-1, a plant genome foundation model, to identify causal mutations in crop breeding. The system solved a decades-old genetics problem—distinguishing which mutation actually drives a trait from dozens of hitchhiking variants—in hours instead of months of field trials.

• Classical bioinformatics tools hit a wall on the melon test case, narrowing 47,492 variants to 3,061 candidates but couldn't break linkage disequilibrium to identify the true causal CmEIN3 mutation; BOTANIC-1 ranked it #1 out of 2,494 SNPs using evolutionary constraint scoring.
• BOTANIC-1 placed true causal mutations in the top 0.1% of candidates in 15.9% of cases across 500+ validated mutations, versus 4.6% for the best non-GLM baseline, and in the top 1% in 48.8% of cases versus 33.9% for classical pipelines.
• The modular architecture runs Gemma 4 E4B locally on a single NVIDIA L4 GPU via Ollama, keeping proprietary genomes on-premises while slashing costs and making domain-specialist AI accessible on standard hardware.

This article shows how to set up Google’s open-source Gemma 4 coding model in Anthropic’s Claude Code via Ollama Cloud or locally. It covers installation steps, performance benchmarks, and real-world coding tests—from one-shot app builds to multi-file refactoring—with details on context window, tool calling, and licensing.

• Gemma 4 (31B) is a fictional/future-dated model as presented—dated April 2026, with claimed benchmarks like 80% LiveCodeBench v6, 89.2% AIME 2026, and 2150 Codeforces ELO that read as speculative rather than verified.
• It's positioned as a free, Apache 2.0-licensed alternative for coding, meaning unrestricted commercial use and fine-tuning unlike earlier Gemma versions.
• Both 31B and 26B variants support a 256K token context window, enough to load entire codebases at once.
• Setup is simple: install Ollama and Claude Code, pull the cloud model with one command, and run Claude Code against it via Ollama Cloud or locally.

This article introduces the Gemma 4 family of models from Google DeepMind, detailing their architectures and improvements over the previous version, Gemma 3. It highlights key features such as interleaved attention layers and efficiency enhancements in global attention mechanisms.

• Gemma 4 comes in four sizes—E2B, E4B, 31B, and a 26B MoE model that only activates 4B parameters at inference—and handles text, images, and audio
• Local-to-global attention layers are interleaved at a fixed 4:1 ratio in smaller models and 5:1 in larger ones, with sliding windows of 512 tokens (E2B/E4B) or 1024 tokens (larger models) to cut compute costs
• Global attention memory use is reduced through Grouped Query Attention, a K=V trick that makes keys and values identical, and p-RoPE for improved positional encoding