Gemma 4 E4B
Released 2026-04 · chat · 128K tokens · 5 benchmarks
Editorial notes
Variante edge/on-device: 8B total / 4.5B effective. Multimodal nativo (texto, imagen, video, audio). Per-Layer Embeddings. Codeforces ELO 940. CoVoST audio 35.54.
Spec sheet
- Company
- Google DeepMind
- Country
- US
- Type
- chat
- Release
- 2026-04
- Context
- 128K tokens
- Slug
- gemma-4-e4b
Benchmarks (5)
Reasoning 3
Cite this model
BibTeX · APA
BibTeX
@misc{frontier-gemma-4-e4b,
title = {Gemma 4 E4B},
author = {{Google DeepMind}},
year = {2026},
note = {Frontier Benchmarks AI atlas. Accessed 2026-05-08},
url = {https://frontierbenchmarks.com/models/gemma-4-e4b}
} APA
Google DeepMind (2026). Gemma 4 E4B [Large language model]. Frontier Benchmarks AI. Retrieved 2026-05-08, from https://frontierbenchmarks.com/models/gemma-4-e4b
Citation reflects the atlas page, not the original model paper. For the paper, see the "Resources" section above.