Home / Models / Gemma 4 26B A4B
Google

Gemma 4 26B A4B

Gemma 4 26B A4B is an open-weights 26B-parameter model by Google. It holds an Intelligence score of 62.8 on the local.ai leaderboard (2026-08 snapshot).

MoEopen weights
62.8
local.ai Intelligence
Parameters
26B
MoE · 4B active parameters
Benchmarks (local.ai snapshot)
τ²-bench
80%
GAIA
65%
GDPval
58%
HuggingFace metadata · updated 2026-07-20
repo
google/gemma-4-26B-A4B-it
license
apache-2.0
downloads
10,168,554
likes
1,391
Official HuggingFace repoOllamaCompare on the savings tool

Metadata refreshed from the HuggingFace API (snapshot 2026-08-15T15:53:42.405Z). Context length and long-form descriptions are the next pipeline stage.

Does it fit your hardware?

Detect or select your GPU first. Green = weights + context headroom, yellow = tight, grey = does not fit.

Q4_K_M 15.1GQ6_K 23.4GQ8_0 28.6GFP16 52G
Quantization≈ File sizeNote
IQ2_XXS7 GBExtreme compression, largest quality loss
IQ3_XXS9.4 GBVery small
Q3_K_M12 GBSmall
NVFP413 GBNVIDIA 4-bit
Q4_K_S14 GBCommon
Q4_K_M15.1 GBMost common sweet spot
Q5_K_M19.8 GBBalanced
Q6_K23.4 GBNear-lossless
Q8_028.6 GBNear-lossless
FP1652 GBOriginal weights

Actual files on HuggingFace

FileSize
model-00001-of-00002.safetensors46.48 GB
model-00002-of-00002.safetensors1.59 GB
GGUF: unsloth/gemma-4-26B-A4B-it-GGUF · 1,393,778 downloads
FileSize
BF16/gemma-4-26B-A4B-it-BF16-00001-of-00002.gguf46.49 GBdownload
BF16/gemma-4-26B-A4B-it-BF16-00002-of-00002.gguf0.54 GBdownload
MTP/mtp-gemma-4-26B-A4B-it-BF16.gguf0.80 GBdownload
MTP/mtp-gemma-4-26B-A4B-it-F16.gguf0.80 GBdownload
MTP/mtp-gemma-4-26B-A4B-it-Q8_0.gguf0.43 GBdownload
gemma-4-26B-A4B-it-MXFP4_MOE.gguf15.41 GBdownload
gemma-4-26B-A4B-it-Q8_0.gguf25.02 GBdownload
gemma-4-26B-A4B-it-UD-IQ2_M.gguf9.33 GBdownload
gemma-4-26B-A4B-it-UD-IQ2_XXS.gguf9.24 GBdownload
gemma-4-26B-A4B-it-UD-IQ3_S.gguf10.51 GBdownload
gemma-4-26B-A4B-it-UD-IQ3_XXS.gguf10.63 GBdownload
gemma-4-26B-A4B-it-UD-IQ4_NL.gguf12.68 GBdownload
GGUF: unsloth/gemma-4-26B-A4B-it-qat-GGUF · 469,680 downloads
FileSize
MTP/mtp-gemma-4-26B-A4B-it-BF16.gguf0.80 GBdownload
MTP/mtp-gemma-4-26B-A4B-it-F16.gguf0.80 GBdownload
MTP/mtp-gemma-4-26B-A4B-it-Q4_0.gguf0.23 GBdownload
MTP/mtp-gemma-4-26B-A4B-it-Q8_0.gguf0.43 GBdownload
gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf13.27 GBdownload
mmproj-BF16.gguf1.11 GBdownload
mmproj-F16.gguf1.11 GBdownload
mmproj-F32.gguf2.13 GBdownload
mtp-gemma-4-26B-A4B-it.gguf0.23 GBdownload

What you save self-hosting this model

Uses the model's own API price when it exists; otherwise the closest commercial equivalent.

Waiting for hardware detection…

Related models

Google
Gemma 4 31B It
31B params · Intelligence 66.2
Google
Gemma 4 12B It
12B params · Intelligence 54
Google
Diffusiongemma 26B A4B It
26B params · Intelligence 50.3
Google
Gemma 4 E4B It
4B params · Intelligence 28.4

RunLocal presents third-party leaderboard data (local.ai / Exo Labs, 2026-08) with its own fit and cost estimates. Verify pricing, licenses and file sizes at the source before deploying a model.