CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.gemma / gemma-4-26B-A4B-it
From: google <huggingface.co/google>Newsgroups: alt.freeweights.gemmaSubject:

gemma-4-26B-A4B-it

Date: 2026-11-14Commit: 4d7ae49Signed: 2DC28CC7B26A283B ✓

Image text to text model with 25.8B parameters. Safetensors weights, 51.6 GB in all. Published by google under Apache-2.0.

25.8B params51.6GBgemma-4-26B-A4B-itsafetensorspermissive
FactsParams25.8BSize51.6 GBFormatssafetensorsTaskimage text to textLicenseApache-2.0From sourcepermissive
$ fw get google/gemma-4-26B-A4B-it
# manifest sha256 1a1d2b1c6d36… · 51.6 GB · 12 files
$ fw verify ./gemma-4-26B-A4B-it google/gemma-4-26B-A4B-it
From sourcegoogle/gemma-4-26B-A4B-itLast check: 2026-11-14 06:30 UTC
Runs withvLLMSGLangFine-tune withAxolotlTRLPEFTUnsloth
OverviewFilesUse itLicenseProvenance

Overview

Size
51.6 GB
Weights
51.6 GB (safetensors)
Parameters
25.8B
Formats
safetensors
Task
image text to text
License
Apache-2.0 (permissive)
Files
12, each with a sha256
Publisher
google
Added
2026-11-14
Section
alt.freeweights.gemma
Lineage

Original release from the publisher.

Base model: google/gemma-4-26B-A4B

Where these bytes come from
Publisher at the pinned commit→You

Sources are tried in order. A file counts only when its size and sha256 match the manifest. Sources are checked when you ask for them, not on a timer.

Sources
BadgeHostLast checkResult
From sourcehuggingface.co2026-11-14 06:30 UTCpassed

Files

12 files · 51.6 GBWeights 51.6 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 12 files with their sha256
FileSizesha256
.eval_results/mmmu_pro.yaml186 Bc310e41b2dfd7a74a89841576cf20bfde94412fda68491d3ceea9ed97f8dcd46
.gitattributes1.71 KBaa4fc74fe06acc698271341dee76eca26c6672a1f4b2840dedbf37f1b4fe27a4
README.md28.0 KB4f463225c4fe49ec120115b0f9bb6713ff1556293a909903ffdae3df7bf974da
chat_template.jinja18.7 KBae53464bf3be25802b3a5b37def7fd89667067d7577049b3b2d74c4d8de4c6d4
config.json3.81 KBed0c1eb3633de771906e9ba004a44cc5635bcc06ee2062077c3d2e88a50707d3
generation_config.json208 Bd4226bbe3117d2d253ba4609720ba82c6c4ce4627a9a6ae05387c78983ac03de
model-00001-of-00002.safetensors49.9 GB1127684971bbca40465435a5cad69d67ad603bf5e61c6dfd5561fae4a3bcfdb3
model-00002-of-00002.safetensors1.70 GBaab47033e1e8a492ef8e581efae1cf36478d0433567e7729b3c1728bc8970db7
model.safetensors.index.json103 KB907826a6e46ff454272bd6db1fee629d5531a2303be22986d825a0871d7dc7a7
processor_config.json1.69 KB32bdf45d2ad4cc29a0822ddd157a182de76644f0419a6228d151495256e9813c
tokenizer.json32.2 MBcc8d3a0ce36466ccc1278bf987df5f71db1719b9ca6b4118264f45cb627bfe0f
tokenizer_config.json3.08 KB9f4fec4b1dc6ecddf8f4a92e9caea5971c0e67d81309f3f9066a2bee8c362633

Use it

$ fw get google/gemma-4-26B-A4B-it
# manifest sha256 1a1d2b1c6d36… · 51.6 GB · 12 files
$ fw verify ./gemma-4-26B-A4B-it google/gemma-4-26B-A4B-it
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/google/gemma-4-26B-A4B-it.json --dir ./gemma-4-26B-A4B-it $ python3 fw.py verify ./gemma-4-26B-A4B-it ./google__gemma-4-26B-A4B-it.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download google/gemma-4-26B-A4B-it --revision 4d7ae4984b7db7de8f8457170b3f1a419ee76d52 --local-dir ./gemma-4-26B-A4B-it
transformersLoad the weights in Python.
import os os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com" from transformers import AutoModelForImageTextToText, AutoProcessor processor = AutoProcessor.from_pretrained("google/gemma-4-26B-A4B-it", revision="4d7ae4984b7db7de8f8457170b3f1a419ee76d52") model = AutoModelForImageTextToText.from_pretrained("google/gemma-4-26B-A4B-it", revision="4d7ae4984b7db7de8f8457170b3f1a419ee76d52", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve google/gemma-4-26B-A4B-it --revision 4d7ae4984b7db7de8f8457170b3f1a419ee76d52
Run in Allternit Desktopcoming in Phase 1
Fine-tune in Freeweights Studiocoming in Phase 3

License

Name
Apache License 2.0
Id
Apache-2.0
Class
Permissive
Redistribution
yes
Text in
README.md
Text sha256
4f463225c4fe49ec120115b0f9bb6713ff1556293a909903ffdae3df7bf974da
Terms gate
No extra terms beyond the license.

Read the license text

Licenses are checked against the license list. A custom license passes only after a recorded human audit.

Provenance

From: google <huggingface.co/google>Newsgroups: alt.freeweights.gemmaSubject: gemma-4-26B-A4B-itDate: 2026-11-14Commit: 4d7ae49Signed: 2DC28CC7B26A283B ✓
Upstream
google/gemma-4-26B-A4B-it
Pinned commit
4d7ae4984b7db7de8f8457170b3f1a419ee76d52
Manifest
gemma-4-26B-A4B-it.json · .minisig
Manifest sha256
1a1d2b1c6d365b123bbaa241dc8a15a9a3f8539d7df1e3bd7ee42552decf0912
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-14 06:30 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/google/gemma-4-26B-A4B-it.json $ curl -O https://freeweights.allternit.com/manifests/google/gemma-4-26B-A4B-it.json.minisig $ minisign -Vm gemma-4-26B-A4B-it.json -P RWQ7KGqyx4zCLXtUFOqvfv8L3g+8fpKJ1FsuxPyOYMF+4KufbUxKq1Oq $ shasum -a 256 <file> # compare with the Files tab

A manifest whose signature fails must be rejected. A file whose size or sha256 does not match must be discarded. Verify a download

Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download google/gemma-4-26B-A4B-it
Run in Allternit Desktopcoming in Phase 1
Signed manifest
minisign, key 2DC28CC7B26A283B. gemma-4-26B-A4B-it.json. Provenance. Takedown history: none.
Source
huggingface.co/google/gemma-4-26B-A4B-it

Related models

All of alt.freeweights.gemma →
From: google <huggingface.co/google>Subject: embeddinggemma-2
From: google <huggingface.co/google>Subject: gemma-4-12B-it
From: google <huggingface.co/google>Subject: gemma-4-31B-it
From: google <huggingface.co/google>Subject: gemma-4-E2B-it
From: google <huggingface.co/google>Subject: gemma-4-E4B-it