{"omc_version":"0.1","id":"google-gemma-4-26b-a4b","name":"Gemma 4 26B A4B","provider":"Google","released":"2026-03-11","licence":"Apache License 2.0","url":"https://huggingface.co/google/gemma-4-26B-A4B-it","description":"Google's Gemma is a mid-size downloadable language model with a permissive Apache licence. It uses an efficient mixture-of-experts design that activates only a few billion parameters per token, making it the efficiency pick of the mid-size class.","modalities":["text","image","video"],"context_window":262144,"capabilities":{"intelligence":3,"writing":3,"coding":3},"x_llmap_capability_sources":{"coding":{"suite":"lmarena-coding","suite_name":"Arena Coding","rank":50,"of":143,"score":1480.7442,"higher_is_better":true,"variant":null},"writing":{"suite":"lmarena-creative-writing","suite_name":"Arena Creative Writing","rank":50,"of":143,"score":1403.9792,"higher_is_better":true,"variant":null},"intelligence":{"suite":"lmarena-text","suite_name":"Arena Text (overall)","rank":47,"of":143,"score":1438.4965,"higher_is_better":true,"variant":null}},"pricing":{"input_per_1m_tokens":0.07,"output_per_1m_tokens":0.34,"currency":"USD","source":"cheapest of 12 tracked offers (OpenRouter)"},"relationships":[{"type":"competes-with","id":"qwen-qwen3-6-27b"},{"type":"competes-with","id":"qwen-qwen3-5-35b-a3b"},{"type":"competes-with","id":"openai-gpt-oss-safeguard-20b"}],"card_meta":{"created":"2026-07-26T07:55:53.048Z","created_by":"LLMap pipeline","last_updated":"2026-08-03T20:14:27.617Z","source":"https://llmap.ai/models/google-gemma-4-26b-a4b","omc_spec":"https://github.com/openmodelcard/spec"}}