CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.olmo / SAGE-MM-Qwen3-VL-4B-SFT_RL
From: allenai <huggingface.co/allenai>Newsgroups: alt.freeweights.olmoSubject:

SAGE-MM-Qwen3-VL-4B-SFT_RL

Date: 2026-11-15Commit: ea4dcedSigned: 2DC28CC7B26A283B ✓

Video text to text model with 4.8B parameters. Safetensors weights, 19.3 GB in all. Published by allenai under Apache-2.0.

4.8B params19.3GBSAGE-MM-Qwen3-VL-4B-SFT_RLsafetensorspermissive
FactsParams4.8BSize19.3 GBFormatssafetensorsTaskvideo text to textLicenseApache-2.0From sourcepermissive
$ fw get allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL
# manifest sha256 35cefca7477d… · 19.3 GB · 19 files
$ fw verify ./SAGE-MM-Qwen3-VL-4B-SFT_RL allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL
From sourceHosted (signed)allenai/SAGE-MM-Qwen3-VL-4B-SFT_RLLast check: 2026-11-15 03:39 UTC
Run at homeGet help

Get help: custom fine-tune or private deploy. We review your setup before we quote.

Fine-tune of SAGE-MM-Qwen3-VL-4B-SFT.

Works with

Run8Fine-tune7Convert8Scan2

Runs with

Ranked by fit to this model (safetensors), then by GitHub stars.

TransformersLocalAItinygradTriton Inference ServerBentoMLCTranslate2ExLlamaV2LitServe
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
19.3 GB
Weights
19.3 GB (safetensors)
Parameters
4.8B
Formats
safetensors
Task
video text to text
License
Apache-2.0 (permissive)
Files
19, each with a sha256
Publisher
allenai
Added
2026-11-15
Section
alt.freeweights.olmo
Lineage

Original release from the publisher.

Base model: allenai/SAGE-MM-Qwen3-VL-4B-SFT

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.co2026-11-15 03:39 UTCpassed

Model card

From the publisher at commit ea4dced. Shown as plain text with links; images open at the publisher.

Image: SAGE Teaser

System Capabilities#

SAGE-MM operates as the core decision-maker within the SAGE system. It functions in two distinct stages:

  1. Stage-1 (Context VLM): The model analyzes initial sampled frames and metadata to determine if the query can be answered immediately ("single-turn") or if it requires tool usage ("multi-turn").
  2. Stage-2 (Iterative Reasoner): If tools are needed, the model enters a loop where it calls tools, analyzes their output, and updates its context until a final answer is derived.
Supported Tools#

The model is trained to generate JSON-formatted actions to invoke the following tools:

  • web-search: Search the internet for external knowledge (e.g., sports standings, cast lists).
  • transcribe-speech: Perform ASR on specific timestamped segments of the video.
  • ground-event: Locate start/end timestamps for specific visual events.
  • extract-video-parts: Extract high-resolution frames or subclips from specific timestamps.
  • analyze: Perform detailed visual analysis on extracted media.

Usage#

Note: SAGE-MM outputs JSON action strings. It requires a runtime environment (provided in our GitHub repo) to parse these strings, execute the tools, and feed the observation back to the model.

License#

This model is licensed under Apache 2.0. It is intended for research and educational use in accordance with Ai2's Responsible Use Guidelines.

Files

19 files · 19.3 GBWeights 19.3 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 19 files with their sha256
FileSizesha256
.gitattributes1.57 KB34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930
README.md1.98 KB7fca05807eb6b0247d1e798a011981be37e39fb8c58144c89ddaf3720f2dabb4
added_tokens.json707 Bc0284b582e14987fbd3d5a2cb2bd139084371ed9acbae488829a1c900833c680
chat_template.jinja5.29 KB3636d0f0bd6bef02654cdffdc447b79cb2cef8ab02cc75267345946291a489e4
config.json1.62 KB2ffac513911d6ec672ccd4038d31750ae4cfab8baa12cd7e51a6e95948a460c8
generation_config.json199 Ba0acd9cf47909f86b54951dbf2cfd2af3b0b36178211926760a3a27edb49378d
merges.txt1.67 MB8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5
metadata.txt272 B5624efcc05807e40168e400d27431eb024f1fcc1c6aabca2fc32bedc4d771dcf
model-00001-of-00004.safetensors4.94 GB2b9f95fd82861b9faf58503fbef2a1872c024f12a9b7b431be6a9802e8faaafe
model-00002-of-00004.safetensors4.94 GBe3253bb533ab337aad4968e80e79051bda150fd7edaa364774c22d1c5b55748a
model-00003-of-00004.safetensors4.94 GB985a9a1f54d10286400bb214c3fb09300b92aba9bd9929653fa0c6e073f15b27
model-00004-of-00004.safetensors4.48 GBa85ec38c39b82b60ef6d007a4491a76fa15bbe668c4839472dc1fc31b4486624
model.safetensors.index.json64.8 KBadc31da53c8add3dcb7993931ecc4bbec71cfc719da837af43a85bdad3dbc50a
preprocessor_config.json782 B93585062a80db5e8ca038efc7726a3e6411d9db948472d81d63c6303993be8c5
special_tokens_map.json613 B76862e765266b85aa9459767e33cbaf13970f327a0e88d1c65846c2ddd3a1ecd
tokenizer.json11.4 MBf54b55fa0c3aba0c91ce09ea79ae4e62da24e2a1a630f96c4bae34aba25e234a
tokenizer_config.json5.55 KBd3ada4329b435d18a4b600e77611a5018f8b13959973c173b2e668fbb7f64027
video_preprocessor_config.json817 B59c5c9eb52182eb14c06ffb10ca9effd29adce5f238a95de23ca14a38dbd2cb1
vocab.json2.78 MBca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
ea4dced0aa pinned now2026-11-15 03:39 UTC19 files · 19.3 GB

Use it

$ fw get allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL
# manifest sha256 35cefca7477d… · 19.3 GB · 19 files
$ fw verify ./SAGE-MM-Qwen3-VL-4B-SFT_RL allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL.json --dir ./SAGE-MM-Qwen3-VL-4B-SFT_RL $ python3 fw.py verify ./SAGE-MM-Qwen3-VL-4B-SFT_RL ./allenai__SAGE-MM-Qwen3-VL-4B-SFT_RL.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL --revision ea4dced0aa91ae86c475279a8e4da82c9beddb3f --local-dir ./SAGE-MM-Qwen3-VL-4B-SFT_RL
Run at home

License

Name
Apache License 2.0
Id
Apache-2.0
Class
Permissive
Redistribution
yes
Listing
Hosted (signed)
Decided
License as declared by the publisher. No person read it. 2026-11-15
Text in
README.md
Text sha256
7fca05807eb6b0247d1e798a011981be37e39fb8c58144c89ddaf3720f2dabb4
Terms gate
No extra terms beyond the license.

Read the license text

Licenses are checked against the license table. A custom license passes only after a person reads it.

Provenance

From: allenai <huggingface.co/allenai>Newsgroups: alt.freeweights.olmoSubject: SAGE-MM-Qwen3-VL-4B-SFT_RLDate: 2026-11-15Commit: ea4dcedSigned: 2DC28CC7B26A283B ✓
Upstream
allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL
Pinned commit
ea4dced0aa91ae86c475279a8e4da82c9beddb3f
Manifest
SAGE-MM-Qwen3-VL-4B-SFT_RL.json · .minisig
Manifest sha256
35cefca7477d4bdf5d426c8fdbbe4439e1de7252f0fd992683b9b8315804e753
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 03:39 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL.json $ curl -O https://freeweights.allternit.com/manifests/allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL.json.minisig $ minisign -Vm SAGE-MM-Qwen3-VL-4B-SFT_RL.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

Model tree

Parent

allenai/SAGE-MM-Qwen3-VL-4B-SFT · this model is a finetune of it

No model in the catalog names this one as its parent.

Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL
Open in Freeweights Studio

Pull, verify and run it in the Allternit app on Desktop, web or phone.

Signed manifest
minisign, key 2DC28CC7B26A283B. SAGE-MM-Qwen3-VL-4B-SFT_RL.json. Provenance. Takedown history: none.
Source
huggingface.co/allenai/SAGE-MM-Qwen3-VL-4B-SFT_RL

Related models

All of alt.freeweights.olmo →
From: bartowski <huggingface.co/bartowski>Subject: allenai_Olmo-3-7B-Instruct-GGUF
From: bartowski <huggingface.co/bartowski>Subject: allenai_Olmo-3-7B-Think-GGUF
From: bartowski <huggingface.co/bartowski>Subject: allenai_Olmo-3.1-32B-Instruct-GGUF
From: bartowski <huggingface.co/bartowski>Subject: allenai_Olmo-3.1-32B-Think-GGUF
From: bartowski <huggingface.co/bartowski>Subject: allenai_olmOCR-2-7B-1025-GGUF
From: bartowski <huggingface.co/bartowski>Subject: allenai_olmOCR-7B-0225-preview-GGUF