HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Models/google/gemma-4-26B-A4B-it

gemma-4-26B-A4B-it

google/gemma-4-26B-A4B-it

kind
original
quant
bf16 (16-bit)
family
gemma
modality
image-text-to-text
root model
google/gemma-4-26B-A4B
provider
google
license
apache-2.0
language
—
context
128K
library
transformers
vRAM (GB)
—
size on disk
—
downloads
13,092,825
likes
1,329
created
—
last modified
2026-07-20T16:42:12+00:00 (2026)
huggingface
https://huggingface.co/google/gemma-4-26B-A4B-it
tags
transformerssafetensorsgemma4image-text-to-textconversationalarxiv:2607.02770base_model:google/gemma-4-26B-A4Bbase_model:finetune:google/gemma-4-26B-A4Blicense:apache-2.0eval-resultsendpoints_compatibledeploy:azureregion:usdeploy:sagemaker

Benchmark scores (10)

benchmarkcategoryscorecontextconfevidence
AIME 2026math89.2AIME 2026 no toolsmedium| AIME 2026 no tools | 89.2% | 88.3% | 77.5% | 42.5% | 37.5% | 20.8% |
MATHmath85.6MATH-Visionmedium| MATH-Vision | 85.6% | 82.4% | 79.7% | 59.5% | 52.4% | 46.0% |
MMLU-Proknowledge85.2MMLU Promedium| MMLU Pro | 85.2% | 82.6% | 77.2% | 69.4% | 60.0% | 67.6% |
GPQA Diamondreasoning84.3GPQA Diamondmedium| GPQA Diamond | 84.3% | 82.3% | 78.8% | 58.6% | 43.4% | 42.4% |
LiveCodeBenchcoding80.0LiveCodeBench v6medium| LiveCodeBench v6 | 80.0% | 77.1% | 72.0% | 52.0% | 44.0% | 29.1% |
MMMUmultimodal76.9MMMU Promedium| MMMU Pro | 76.9% | 73.8% | 69.1% | 52.6% | 44.2% | 49.7% |
BigBenchreasoning74.4BigBench Extra Hardmedium| BigBench Extra Hard | 74.4% | 64.8% | 53.0% | 33.1% | 21.9% | 19.3% |
Humanity's Last Examknowledge26.5HLE with searchmedium| HLE with search | 26.5% | 17.2% | - | - | - | - |
Humanity's Last Examknowledge19.5HLE no toolsmedium| HLE no tools | 19.5% | 8.7% | 5.2% | - | - | - |
WER FLEURSspeech-asr8.0FLEURS (lower is better)medium| FLEURS (lower is better) | - | - | 0.069<sup>*</sup> | 0.08 | 0.09 | - |

Discussions (53)

titleauthorstatuscommentstype
[Question] update in templatemsimunekopen0disc
Update READMEMaartenGrmerged0disc
Raise per-image vision soft-token budget from 280 to 1120lucianommartinsopen0disc
Add response_template to tokenizer_config.jsonRocketknight1merged0disc
Multilingual powerhouse — testing for mobile deployment3morixdopen0disc
Chat template may re-inject prior-turn reasoning during multi-turn tool use → repetition loopsManniX-ITAopen0disc
fix: chat template — null handling, reasoning preservation, turn-tag balance, input validationlucianommartinsmerged0disc
Add EvalEval community eval resultsEvalEvalBotclosed0disc
fun classification modelbaseten-adminopen0disc
Add MMMU-Pro evaluation resultSaylorTwiftmerged0disc
[Showcase] Running Gemma-4-26B-A4B-it on 8GB RAM SmartphoneInfiniteVoidopen0disc
Fix empty thought channel in Gemma 4 chat templatedutifulbobclosed0disc
Fix lineageosansevieromerged0disc
Fix: Chat_template.jinja - Add Generation promptilaytopen0disc
QAT?sbeltzopen0disc
fix(chat_template): Emit multimodal placeholders in tool response content-partsharshaljanjanimerged0disc
GGUF available — Cerebellum v3 (11 GB, ablation-guided mixed-precision)deucebucketopen0disc
fix(chat_template): update SI and tool call handlingdougreidmerged0disc
add newlines and thinking tokens to template to avoid having to compute 3 extra tokens per generation in chat completion+reasoningquasar-of-mikusopen0disc
Very bad results with model quant and KV cache quant, only BF16 works wellqenmeopen0disc
Fix chat_template: emit empty <|channel>thought\n<channel|> wrapper for existing asst turnsflotherxiopen0disc
[Bug] chat_template: missing <|channel>thought\n<channel|> wrapper for non-thinking SFT / multi-turnflotherxiopen0disc
Release the 124B parent weights... We know you have it.Durekaopen0disc
tfhe_ntt::prime32::Plan::try_newmilezdeep13draft0disc
Add ParseBench evaluation resultsboyang-runllamaopen0disc
Add ParseBench evaluation resultsboyang-runllamaclosed0disc
Thinking Mode doesn't work properly on gemma-4-26B-A4B-it.michaelkopf1981open0disc
Fix missing thinking channel in Gemma 4 chat template when using continue_final_messageCalinRopen0disc
Your 260k dictionary is breaking Gemma 4's back.phil111open0disc
fix: embed chat_template in tokenizer_config.jsonNERDDISCOopen0disc
How to download a model like git lfs?PYTHON01100100closed0disc
Update response schemaRocketknight1merged0disc
fix: prefix-preserving tool calls + dialog compliancelucianommartinsmerged0disc
fix: function calling formatting in chat templateRyanMullinsopen0disc
Vertex AI & vLLM Deployment Guide for Gemma 4 26B-A4B-it (MoE) + Known LimitationsManzela-Dopen0disc
Thank you Google!KngRnZopen0disc
[Appreciation] Incredible performance of Gemma 4-26b on consumer hardware — 90 t/s even on an older DDR3 system!MightyLoraLordopen0disc
Excellent release, Google. Gemma 4 is good.DorkMckork1open0disc
Fantastic release!Dampfinchenopen0disc
Significant Otter! ❤MrDevolveropen0disc
THANK YOU! GoogleE7Reineopen0disc
gemmaMalay17bdraft0disc
Are you guys going to add other MoE stuff?nesymerp1open0disc
Gemma 4, Compared to G3, Has Notable Improvements And Regressionsphil111closed0disc
Verified Commit?stephenrawlsopen0disc
please fp8huang123chuanopen0disc
First community NVFP4 quantization of Gemma 4 26B-A4B-it (49GB → 16.5GB)marioiseliopen0disc
Good Work. Please consider addressing factual hallucinations.phil111closed0disc
Add AIME 2026 evaluation resultSaylorTwiftopen0disc
ordl-gemma-familyWinsocketclosed0disc
add eval resultsmerveopen0disc
Proposed video configurationpcuenqmerged0disc
error when batch size >1loulou2open0disc
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.