CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.design

One card, four faces

Every model has one card. It has four faces that share the same data and the same dot lettering. Identity is the element tile, Facts is the spec sheet, Record is the newsgroup header and Action is the terminal. Each view shows the faces that fit it.

IdentityGrid front, home, related models. Monogram, size, publisher logo.
FactsGrid back (hover, tap or Enter flips it). Next to Identity in the deck and on the model page.
RecordThe list rows, the strip under each tile, the top of every combined card and the Provenance tab.
ActionThe foot of the combined card, the Use it tab and fw.py info in the terminal.

Anatomy, desktop

From: Qwen <huggingface.co/Qwen>Newsgroups: alt.freeweights.qwenSubject: Qwen3.6-27BDate: 2026-11-14Commit: 8f33b98Signed: 2DC28CC7B26A283B ✓

Image text to text model with 27.8B parameters. Safetensors weights, 55.6 GB in all. Published by Qwen under Apache-2.0.

27.8B params55.6GBQwen3.6-27Bsafetensorspermissive
FactsParams27.8BSize55.6 GBFormatssafetensorsTaskimage text to textLicenseApache-2.0From sourcepermissive
$ fw get Qwen/Qwen3.6-27B
# manifest sha256 d66d703e403a… · 55.6 GB · 29 files
$ fw verify ./Qwen3.6-27B Qwen/Qwen3.6-27B
Open model page
From: openai <huggingface.co/openai>Newsgroups: alt.freeweights.speechSubject: whisper-large-v3-turboDate: 2026-11-14Commit: 41f01f3Signed: 2DC28CC7B26A283B ✓

Automatic speech recognition model with 809M parameters. Safetensors weights, 1.62 GB in all. Published by openai under MIT.

809M params1.62GBwhisper-large-v3-turbosafetensorspermissive
FactsParams809MSize1.62 GBFormatssafetensorsTaskautomatic speech recognitionLicenseMITFrom sourcepermissive
$ fw get openai/whisper-large-v3-turbo
# manifest sha256 89fd30054197… · 1.62 GB · 13 files
$ fw verify ./whisper-large-v3-turbo openai/whisper-large-v3-turbo
Open model page
From: black-forest-labs <huggingface.co/black-forest-labs>Newsgroups: alt.freeweights.imageSubject: flux-3-image-klein-comfyDate: 2026-11-14

Image to image model. Safetensors weights, 93.7 GB in all. Published by black-forest-labs under FLUX-Kommunity-1.0 (custom).

image93.7GBflux-3-image-klein-comfysafetensorsgated
FactsParamsnot statedSize93.7 GBFormatssafetensorsTaskimage to imageLicenseFLUX-Kommunity-1.0 (custom)From the publishergatedread terms
$ fw info black-forest-labs/flux-3-image-klein-comfy
link only — read the license:
https://huggingface.co/black-forest-labs/flux-3-image-klein-comfy/blob/9d36114cf7a2e03a45c28f5cc4987d20ebfdb6c1/LICENSE.md
License and source

Phone, 390 px

From: Qwen <huggingface.co/Qwen>Newsgroups: alt.freeweights.qwenSubject: Qwen3.6-27BDate: 2026-11-14Commit: 8f33b98Signed: 2DC28CC7B26A283B ✓

Image text to text model with 27.8B parameters. Safetensors weights, 55.6 GB in all. Published by Qwen under Apache-2.0.

27.8B params55.6GBQwen3.6-27Bsafetensorspermissive
FactsParams27.8BSize55.6 GBFormatssafetensorsTaskimage text to textLicenseApache-2.0From sourcepermissive
$ fw get Qwen/Qwen3.6-27B
# manifest sha256 d66d703e403a… · 55.6 GB · 29 files
$ fw verify ./Qwen3.6-27B Qwen/Qwen3.6-27B
Open model page
From: Qwen <huggingface.co/Qwen>Subject: Qwen3.6-27B
From: openai <huggingface.co/openai>Newsgroups: alt.freeweights.speechSubject: whisper-large-v3-turboDate: 2026-11-14Commit: 41f01f3Signed: 2DC28CC7B26A283B ✓

Automatic speech recognition model with 809M parameters. Safetensors weights, 1.62 GB in all. Published by openai under MIT.

809M params1.62GBwhisper-large-v3-turbosafetensorspermissive
FactsParams809MSize1.62 GBFormatssafetensorsTaskautomatic speech recognitionLicenseMITFrom sourcepermissive
$ fw get openai/whisper-large-v3-turbo
# manifest sha256 89fd30054197… · 1.62 GB · 13 files
$ fw verify ./whisper-large-v3-turbo openai/whisper-large-v3-turbo
Open model page
From: openai <huggingface.co/openai>Subject: whisper-large-v3-turbo
From: black-forest-labs <huggingface.co/black-forest-labs>Newsgroups: alt.freeweights.imageSubject: flux-3-image-klein-comfyDate: 2026-11-14

Image to image model. Safetensors weights, 93.7 GB in all. Published by black-forest-labs under FLUX-Kommunity-1.0 (custom).

image93.7GBflux-3-image-klein-comfysafetensorsgated
FactsParamsnot statedSize93.7 GBFormatssafetensorsTaskimage to imageLicenseFLUX-Kommunity-1.0 (custom)From the publishergatedread terms
$ fw info black-forest-labs/flux-3-image-klein-comfy
link only — read the license:
https://huggingface.co/black-forest-labs/flux-3-image-klein-comfy/blob/9d36114cf7a2e03a45c28f5cc4987d20ebfdb6c1/LICENSE.md
License and source
From: black-forest-labs <huggingface.co/black-forest-labs>Subject: flux-3-image-klein-comfy

Tools and datasets

Every catalog entry is built the same way. A tool shows its latest release where a model shows its size, and its Action face is the install command. A dataset's Action face is load_dataset.

From: ggml-org <github.com/ggml-org>Newsgroups: alt.freeweights.toolsSubject: llama.cppRelease: v1.0.0 (2026-11-13)Commit: b1b770dLicense: MIT

Runs GGUF language models on CPUs and consumer GPUs, with a local server.

inferencev1.0.0releasellama.cppC++MIT
FactsLicenseMITLanguageC++KindinferenceRelease2026-11-13PlatformsmacOS, Linux, WindowsLinks to the projectpermissive
$ brew install llama.cpp
# use with Freeweights
$ llama-server -m ./<file>.gguf --port 8080
Open model page
From: HuggingFaceFW <huggingface.co/HuggingFaceFW>Newsgroups: alt.freeweights.datasetsSubject: fineweb-eduDate: 2026-11-14Commit: 87f0914Signed: 2DC28CC7B26A283B ✓

Dataset. Parquet files, 28.5 GB in all. Published by HuggingFaceFW under ODC-BY-1.0.

datasets28.5GBfineweb-eduparquetpermissive
FactsRowsnot statedSize28.5 GBFormatparquetLicenseODC-BY-1.0From sourcepermissive
>>> from datasets import load_dataset
>>> ds = load_dataset("HuggingFaceFW/fineweb-edu")
# with HF_ENDPOINT=https://freeweights.allternit.com · manifest sha256 17418e09123c…
Open model page
From: Comfy-Org <github.com/Comfy-Org>Subject: ComfyUI
From: ggml-org <github.com/ggml-org>Subject: llama.cpp
From: ggml-org <github.com/ggml-org>Subject: whisper.cpp
From: HuggingFaceFW <huggingface.co/HuggingFaceFW>Subject: fineweb-edu
From: Idavidrein <huggingface.co/Idavidrein>Subject: gpqa

In the terminal

The same card in fw.py info and fw.py search: Record lines, key facts and the Action commands. Color only on a terminal, never when NO_COLOR is set. fw.py reference

┌─ alt.freeweights.qwen ────────────────────────────────────── fw ─┐
│ ⠀⣿⣿⣿⠀⠀⣿⠀⠀⠀⣿   From:       Qwen <huggingface.co/Qwen>             │
│ ⣿⠀⠀⠀⣿⠀⣿⠀⠀⠀⣿   Newsgroups: alt.freeweights.qwen                   │
│ ⣿⠀⣿⠀⣿⠀⣿⠀⣿⠀⣿   Subject:    Qwen3.6-27B                            │
│ ⣿⠀⠀⣿⠀⠀⣿⣿⠀⣿⣿   Date:       2026-11-14                             │
│ ⠀⣿⣿⠀⣿⠀⣿⠀⠀⠀⣿   Commit:     8f33b98                                │
│               Signed:     2DC28CC7B26A283B ✓                     │
├──────────────────────────────────────────────────────────────────┤
│ Image text to text model with 27.8B parameters. Safetensors,     │
│ 55.6 GB in all. Published by Qwen under Apache-2.0.              │
├──────────────────────────────────────────────────────────────────┤
│ params 27.8B  ·  size 55.6 GB  ·  safetensors  ·  image text to  │
│ text  ·  Apache-2.0 (permissive)                                 │
├──────────────────────────────────────────────────────────────────┤
│ $ fw get Qwen/Qwen3.6-27B                                        │
│ # manifest sha256 d66d703e403a… · 29 files                       │
│ $ fw verify ./Qwen3.6-27B Qwen/Qwen3.6-27B                       │
└──────────────────────────────────────────────────────────────────┘

The four original styles

Kept as references. Each one became a face of the card above.

A Spec sheet

Now the FACTS face. Params, size, formats, license class, source badge and the terms flag.

Desktop
alt.freeweights.qwen
Qwen3.6-27B
Qwen/Qwen3.6-27B
Params
27.8B
Size
55.6 GB
Format
safetensors
License
Apache-2.0
From sourcepermissiveimage text to text
alt.freeweights.speech
whisper-large-v3-turbo
openai/whisper-large-v3-turbo
Params
809M
Size
1.62 GB
Format
safetensors
License
MIT
From sourcepermissiveautomatic speech recognition
alt.freeweights.image
flux-3-image-klein-comfy
black-forest-labs/flux-3-image-klein-comfy
Params
not stated
Size
93.7 GB
Format
safetensors
License
FLUX-Kommunity-1.0 (custom)
From the publishergatedread termsimage to image
Phone, 390 px
alt.freeweights.qwen
Qwen3.6-27B
Qwen/Qwen3.6-27B
Params
27.8B
Size
55.6 GB
Format
safetensors
License
Apache-2.0
From sourcepermissiveimage text to text
alt.freeweights.speech
whisper-large-v3-turbo
openai/whisper-large-v3-turbo
Params
809M
Size
1.62 GB
Format
safetensors
License
MIT
From sourcepermissiveautomatic speech recognition

B Newsgroup post

Now the RECORD face. From, Newsgroups, Subject, Date, Commit and Signed lines in mono, then a one-line summary.

Desktop
From: Qwen <huggingface.co/Qwen>
Newsgroups: alt.freeweights.qwen
Subject: Qwen3.6-27B (55.6 GB)
Date: Sat, 14 Nov 2026

Image text to text model with 27.8B parameters. Safetensors weights, 55.6 GB in all. Published by Qwen under Apache-2.0.

--
Apache-2.0 · permissive · from source
From: openai <huggingface.co/openai>
Newsgroups: alt.freeweights.speech
Subject: whisper-large-v3-turbo (1.62 GB)
Date: Sat, 14 Nov 2026

Automatic speech recognition model with 809M parameters. Safetensors weights, 1.62 GB in all. Published by openai under MIT.

--
MIT · permissive · from source
From: black-forest-labs <huggingface.co/black-forest-labs>
Newsgroups: alt.freeweights.image
Subject: flux-3-image-klein-comfy (93.7 GB)
Date: Sat, 14 Nov 2026

Image to image model. Safetensors weights, 93.7 GB in all. Published by black-forest-labs under FLUX-Kommunity-1.0 (custom).

--
FLUX-Kommunity-1.0 (custom) · gated · from the publisher · read terms first
Phone, 390 px
From: Qwen <huggingface.co/Qwen>
Newsgroups: alt.freeweights.qwen
Subject: Qwen3.6-27B (55.6 GB)
Date: Sat, 14 Nov 2026

Image text to text model with 27.8B parameters. Safetensors weights, 55.6 GB in all. Published by Qwen under Apache-2.0.

--
Apache-2.0 · permissive · from source
From: openai <huggingface.co/openai>
Newsgroups: alt.freeweights.speech
Subject: whisper-large-v3-turbo (1.62 GB)
Date: Sat, 14 Nov 2026

Automatic speech recognition model with 809M parameters. Safetensors weights, 1.62 GB in all. Published by openai under MIT.

--
MIT · permissive · from source

C Element tile

Now the IDENTITY face. The family monogram in dot lettering, a big size figure and a small facts row.

Desktop
27.8Bpermissive
55.6GB
Qwen3.6-27B
safetensorsimage text to text
809Mpermissive
1.62GB
whisper-large-v3-turbo
safetensorsautomatic speech recognition
—gated
93.7GB
flux-3-image-klein-comfy
safetensorsimage to image
Phone, 390 px
27.8Bpermissive
55.6GB
Qwen3.6-27B
safetensorsimage text to text
809Mpermissive
1.62GB
whisper-large-v3-turbo
safetensorsautomatic speech recognition

D Terminal

Now the ACTION face. The fw get command, the manifest sha256 prefix and fw verify. Link-only rows say so instead.

Desktop
alt.freeweights.qwen
$ fw get Qwen/Qwen3.6-27B
# manifest sha256 d66d703e403a…
# 55.6 GB · 29 files · safetensors
# license Apache-2.0 (permissive)
[source] huggingface.co @ 8f33b98
alt.freeweights.speech
$ fw get openai/whisper-large-v3-turbo
# manifest sha256 89fd30054197…
# 1.62 GB · 13 files · safetensors
# license MIT (permissive)
[source] huggingface.co @ 41f01f3
alt.freeweights.image
$ fw info black-forest-labs/flux-3-image-klein-comfy
# link only: FLUX-Kommunity-1.0 (custom) (gated)
# from the publisher, never hosted here
[publisher] huggingface.co
Phone, 390 px
alt.freeweights.qwen
$ fw get Qwen/Qwen3.6-27B
# manifest sha256 d66d703e403a…
# 55.6 GB · 29 files · safetensors
# license Apache-2.0 (permissive)
[source] huggingface.co @ 8f33b98
alt.freeweights.speech
$ fw get openai/whisper-large-v3-turbo
# manifest sha256 89fd30054197…
# 1.62 GB · 13 files · safetensors
# license MIT (permissive)
[source] huggingface.co @ 41f01f3