Gemma-4 lands in Vision Arena as #2 & #4 open models, and shifts the Pareto frontier!
TL;DR · AI Summary
Gemma-4 models rank second and fourth in the vision domain, significantly shifting the performance boundary of open models.
Key Takeaways
- Gemma-4-31b ranks #2 among open models (overall #20).
- Gemma-4-26b-a4b ranks #4 among open models (overall #26).
- Google DeepMind's models dominate the price-performance Pareto in both proprieta
Outline
Jump quickly between sections.
Gemma-4 models perform exceptionally well among open models in the vision domain.
Gemma-4-31b ranks #2 among open models (overall #20).
Gemma-4-26b-a4b ranks #4 among open models (overall #26).
Google DeepMind's models dominate the price-performance Pareto in both proprietary and open models.
Mindmap
See how the topics connect at a glance.
查看大纲文本（无障碍 / 无 JS 友好）
- Gemma-4模型
- 排名
- Gemma-4-31b
- Gemma-4-26b-a4b
- Google DeepMind
- 价格性能比
Highlights
Key sentences worth saving and sharing.
Gemma-4-31b ranks #2 open (#20 overall), showcasing its leading position among open models.
Gemma-4-26b-a4b ranks #4 open (#26 overall), further consolidating Google DeepMind's technological advantage.
@GoogleDeepMind dominates the price-performance Pareto in Vision across both proprietary and open models.
@GoogleDeepMind dominates the price-performance Pareto in Vision across both proprietary and open models.
- Gemma-4-31b ranks #2 open (#20 overall)
- Gemma-4-26b-a4b ranks #4 open (#26 overall) https://t.co/rieuIlttpH" / X
Arena.ai on X: "Gemma-4 lands in Vision Arena as #2 & #4 open models, and shifts the Pareto frontier! @GoogleDeepMind dominates the price-performance Pareto in Vision across both proprietary and open models. - Gemma-4-31b ranks #2 open (#20 overall) - Gemma-4-26b-a4b ranks #4 open (#26 overall) https://t.co/rieuIlttpH" / X
Don’t miss what’s happening
Gemma-4 lands in Vision Arena as #2 & #4 open models, and shifts the Pareto frontier!
dominates the price-performance Pareto in Vision across both proprietary and open models. - Gemma-4-31b ranks #2 open (#20 overall) - Gemma-4-26b-a4b ranks #4 open (#26 overall) The Vision Arena ranks multimodal AI models capable of reasoning over visual inputs. Congrats to
again on the open model progress!
Quote
Google DeepMind
@GoogleDeepMind
·
Apr 2
Meet Gemma 4: our new family of open models you can run on your own hardware. Built for advanced reasoning and agentic workflows, we’re releasing them under an Apache 2.0 license. Here’s what’s new 
The media could not be played.
Reload
·
7
26
262
49
Read 7 replies