google
/

gemma-4-26B-A4B-it

Image-Text-to-Text

Model card Files Files and versions

Resources

View closed (8)

Release the 124B parent weights... We know you have it.

#31 opened about 10 hours ago by

tfhe_ntt::prime32::Plan::try_new

#30 opened 2 days ago by

Add ParseBench evaluation results

#29 opened 4 days ago by

boyang-runllama

Thinking Mode doesn't work properly on gemma-4-26B-A4B-it.

#27 opened 5 days ago by

michaelkopf1981

Fix missing thinking channel in Gemma 4 chat template when using continue_final_message

#26 opened 6 days ago by

Your 260k dictionary is breaking Gemma 4's back.

#25 opened 7 days ago by

fix: embed chat_template in tokenizer_config.json

#24 opened 8 days ago by

fix: function calling formatting in chat template

#20 opened 12 days ago by

Vertex AI & vLLM Deployment Guide for Gemma 4 26B-A4B-it (MoE) + Known Limitations

#19 opened 12 days ago by

Thank you Google!

#18 opened 13 days ago by

[Appreciation] Incredible performance of Gemma 4-26b on consumer hardware — 90 t/s even on an older DDR3 system!

#17 opened 13 days ago by

Excellent release, Google. Gemma 4 is good.

#16 opened 14 days ago by

Fantastic release!

#15 opened 15 days ago by

Significant Otter! ❤

#14 opened 17 days ago by

THANK YOU! Google

#13 opened 17 days ago by

gemma

#12 opened 17 days ago by

Are you guys going to add other MoE stuff?

#11 opened 17 days ago by

Verified Commit?

#9 opened 18 days ago by

please fp8

#8 opened 18 days ago by

First community NVFP4 quantization of Gemma 4 26B-A4B-it (49GB → 16.5GB)

#7 opened 18 days ago by

Add AIME 2026 evaluation result

#5 opened 19 days ago by

add eval results

#3 opened 19 days ago by

error when batch size >1

#1 opened about 1 month ago by