Google: Gemma 4 26B A4B

Gemma 4 26B A4B is a Google model that accepts text, images, and video as input, making it one of the multimodal options in this tier. It supports a 262,144-token context window, tool use, and reasoning, which covers most workflow requirements outside of structured output (status not confirmed). Pricing sits at $0.06 per million input tokens and $0.33 per million output tokens. On benchmarks, the model scores 51.0 on a blended basis across four tests, with notable variation internally: coding reaches 64.9 while agentic performance sits at 18.1, suggesting a model better suited to code assistance and document analysis than autonomous task execution. Buyers prioritizing low input costs alongside multimodal and long-context support will find it competitive, but those running agentic pipelines should weigh that 18.1 score carefully against alternatives before committing.