Gemma 4 E4B IT
- Context window
- 131K 131,072 tokens
- Input / Output
- text, image, video, audio → text
Timeline
- · Weights version · evidence link
- · Evidence links · GitHub search · Google · pulls · smg-project/smg · @hello-alexmcc
Excerpt
moves nothing. The template differs from Gemma 4's (gemma-4-E4B-it's at the pin) in a Jinja comment block alone, so nothing renders differently. The recorded outputs
- · Evidence links · GitHub search · Google · pulls · smg-project/smg · @hello-alexmcc
Excerpt
dler-llama-nemotron-8b-merge-research 416,850 -> 416,856; dots3-note-prev 589,296 -> 589,308; gemma-4-e4b-it 478,241 -> 478,253; glm-4.6 492,389 -> 492,400; glm-4.7-flash 1,947 -> 1,958; glm-5.3-flash 503,273 -> 503,285; hermes-4-14b
Specifications
Gemma 4 E4B IT is a model from Google with open weights, released on 31 Mar 2026. The context window is 131,072 tokens.
- Context window
- 131,072
- Input
- text · image · video · audio
- Output
- text
- Weights
- Open
Questions
What is the context window of Gemma 4 E4B IT?
The context window is 131,072 tokens.
Are the weights of Gemma 4 E4B IT publicly available?
Yes. Google publishes the weights.
When was Gemma 4 E4B IT released?
On 31 Mar 2026.