CodeGemma 7B

Good fit for

Developers exploring a local natural-language-to-code assistant.

Why consider it

The catalog distinguishes instruction-tuned code chat from separate pretrained code-completion variants; this tag supplies an instruction-style template.

Main limitations

Do not transfer the 2B code variant’s fill-in-the-middle example directly to the instruction-tuned tag. Generated code needs review and testing. The 5GB package is not a memory requirement. No code executor is bundled with the model.

Cost and access

No subscription or per-token hosted fee for local inference within permitted uses. Hardware, electricity and setup are separate. Gemma Terms of Use apply; this older release does not inherit Gemma 4’s Apache license.

Getting started

Install a current compatible Ollama version, review the exact license and run: ollama run codegemma:7b. API refers to the local runtime endpoint, not a hosted API included with the weights.