Gemma 4 E2B
#35 in Open-Source LLMsgoogle · v4 · e2b · since 2026-04-02 · 11× · last seen Jun 30, 2026
14
Momentum
Gemma 4 E2B is a language model by Google with approximately 2.3–2.5 billion effective parameters, optimized for edge devices such as smartphones and Raspberry Pi. It supports multimodality with text and image inputs and can generate text output.
Momentum trend
04.04.03.07.
Features
| Key Benchmark (%) | MMLU-Pro: 60.0% | AIME 2026: 37.5% | LiveCodeBench v6: 44.0% | MMMU Pro (Vision): 44.2% (all values from the official Google DeepMind technical report dated April 2, 2026) |
| Context Window (Tokens) | 128,000 tokens |
| License | Apache 2.0 (unrestricted commercial use) |
| Multimodality | Text + image (variable resolution/aspect ratio) + audio (ASR & speech translation, up to 30s, E2B & E4B only) + video (sequence of frames); output: text |
| Price | Free (open weights, runs locally); no API price from Google for the base model |
| Price per 1M Tokens | Not applicable – open-weights model with no API price; free to run locally |
| Release Date | April 2, 2026 |