CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.vision / ModernBERT-large-zeroshot-v2.0
From: MoritzLaurer <huggingface.co/MoritzLaurer>Newsgroups: alt.freeweights.visionSubject:

ModernBERT-large-zeroshot-v2.0

Date: 2026-11-15Commit: a51e07bSigned: 2DC28CC7B26A283B ✓

Text classification model with 396M parameters. Onnx and safetensors weights, 5.49 GB in all. Published by MoritzLaurer under Apache-2.0.

396M params5.49GBModernBERT-large-zeroshot-v2.0onnx safetensorspermissive
FactsParams396MSize5.49 GBFormatsonnx, safetensorsTasktext classificationLicenseApache-2.0From sourcepermissive
$ fw get MoritzLaurer/ModernBERT-large-zeroshot-v2.0
# manifest sha256 2574394a85d0… · 5.49 GB · 16 files
$ fw verify ./ModernBERT-large-zeroshot-v2.0 MoritzLaurer/ModernBERT-large-zeroshot-v2.0
From sourceHosted (signed)MoritzLaurer/ModernBERT-large-zeroshot-v2.0Last check: 2026-11-15 04:46 UTC
Run at homeGet help

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

Quantized of ModernBERT-large.

Works with

Run10Fine-tune7Convert8Scan2
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
5.49 GB
Weights
792 MB (safetensors)
Parameters
396M
Formats
onnx, safetensors
Task
text classification
License
Apache-2.0 (permissive)
Files
16, each with a sha256
Publisher
MoritzLaurer
Added
2026-11-15
Section
alt.freeweights.vision
Lineage

Original release from the publisher.

Base model: answerdotai/ModernBERT-large

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 04:46 UTCpassed

Model card

From the publisher at commit a51e07b. Shown as plain text with links; images open at the publisher.

ModernBERT-base-zeroshot-v2.0#

Model description#

This model is answerdotai/ModernBERT-large fine-tuned on the same dataset mix as the zeroshot-v2.0 models in the Zeroshot Classifiers Collection.

General takeaways:#

  • The model is very fast and memory efficient. It's multiple times faster and consumes multiple times less memory than DeBERTav3.

The memory efficiency enables larger batch sizes. I got a ~2x speed increase by enabling bf16 (instead of fp16).

  • It performs slightly worse then DeBERTav3 on average on the tasks tested below.
  • I'm in the process of preparing a newer version trained on better synthetic data to make full use of the 8k context window

and to update the training mix of the older zeroshot-v2.0 models.

Training results#
DatasetsMeanMean w/o NLImnli_mmnli_mmfevernlianli_r1anli_r2anli_r3wanlilingnliwellformedqueryrottentomatoesamazonpolarityimdbyelpreviewshatexplainmassivebanking77emotiondairemocontextempatheticagnewsyahootopicsbiasframes_sexbiasframes_offensivebiasframes_intentfinancialphrasebankappreviewshateoffensivetrueteacherspamwikitoxic_toxicaggregatedwikitoxic_obscenewikitoxic_identityhatewikitoxic_threatwikitoxic_insultmanifestocapsotu
Accuracy0.850.8510.9420.9440.8940.8120.7170.7160.8360.9090.8150.8990.9640.9510.9840.8140.80.7440.7520.8020.5440.8990.7350.9340.8640.8770.9130.9530.9210.8210.9890.9010.9270.9310.9590.9110.4970.73
F1 macro0.8340.8350.9350.9380.8820.7950.6880.6760.8230.8980.8140.8990.9640.9510.9840.770.7530.7630.690.8050.5330.8990.7290.9250.8640.8770.9010.9530.8550.8210.9830.9010.9270.9310.9520.9110.3620.662
Inference text/sec (A100 40GB GPU, batch=32)1116.01104.01039.01241.01138.01102.01124.01133.01251.01240.01263.01231.01054.0559.0795.01238.01312.01285.01273.01268.0992.01222.0894.01176.01194.01197.01206.01166.01227.0541.01199.01045.01054.01020.01005.01063.01214.01220.0
Training hyperparameters#

The following hyperparameters were used during training:

  • learning_rate: 9e-06
  • train_batch_size: 16
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.06
  • num_epochs: 2
Framework versions#
  • Transformers 4.48.0.dev0
  • Pytorch 2.5.1+cu124
  • Datasets 3.2.0
  • Tokenizers 0.21.0

Files

16 files · 5.49 GBWeights 792 MB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 16 files with their sha256
FileSizesha256
.gitattributes1.52 KB11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
README.md3.25 KBf5edbc9131c60af9e108be1e2ea2d4edfe75e606218114994059cee65e5f4d1b
config.json1.47 KBababd87e19a2c783e08495996e5e32e7d6661ea66686ba41fd02f9e7d46d547f
model.safetensors792 MB9e88e3ba0721cee770645d262f10dbe8125ed592ff349b7ba2eedaaded97230a
onnx/model.onnx1.58 GBf46163a517e91ce30a1de9a93eaa54b735544d5ad52512a1690e0516a29ac26b
onnx/model_bnb4.onnx401 MB9c84daf7f24185d632285cb2e24ece4019d19618953548e0d47c9f6ee6c80fd9
onnx/model_fp16.onnx792 MBc1f22b94c38e4b8808c31155bc8b9dbc76a2ed94f0279ae39c7f74019e724404
onnx/model_int8.onnx398 MB4fcd879d3433e2fff506ac86221b4656f6a960653752ee87d087efd3a64cc128
onnx/model_q4.onnx422 MBd8a96829d00501f1289c12b6d283d86c423575f1534c99391eb9082fed88cd53
onnx/model_q4f16.onnx297 MB175edda604087bc26c9753eda45e59cd145ebf65101bbceb8f62074a88ff5db1
onnx/model_quantized.onnx398 MB4fcd879d3433e2fff506ac86221b4656f6a960653752ee87d087efd3a64cc128
onnx/model_uint8.onnx398 MB5342485779bc108df404adf594bda01c5cca4399a880c9bc663b90eaaaed595c
special_tokens_map.json694 Bea97ecdbcc73713039d8d64dbb05e3689495c96657fbd9a18f5bed381be81049
tokenizer.json3.58 MB6c8aaa9a542084f2457eab775d4eeb51f92a70c0fd9de28d5edb0ddec3c08d30
tokenizer_config.json20.8 KBc5a7dfd44d93c5d7ed9b2ddcd7a6018e935ec7d53f1de45910ea9868e0ff1c17
training_args.bin5.50 KB61821e0fdeffc4b611be361d1b14a797e9bac02a8df32fbc27cfca5b2aa7b51a
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
a51e07b524 pinned now2026-11-15 04:46 UTC16 files · 5.49 GB

Use it

$ fw get MoritzLaurer/ModernBERT-large-zeroshot-v2.0
# manifest sha256 2574394a85d0… · 5.49 GB · 16 files
$ fw verify ./ModernBERT-large-zeroshot-v2.0 MoritzLaurer/ModernBERT-large-zeroshot-v2.0
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/MoritzLaurer/ModernBERT-large-zeroshot-v2.0.json --dir ./ModernBERT-large-zeroshot-v2.0 $ python3 fw.py verify ./ModernBERT-large-zeroshot-v2.0 ./MoritzLaurer__ModernBERT-large-zeroshot-v2.0.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download MoritzLaurer/ModernBERT-large-zeroshot-v2.0 --revision a51e07b524299e309dd2b88d48b0cfa2bd9ec598 --local-dir ./ModernBERT-large-zeroshot-v2.0
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
f5edbc9131c60af9e108be1e2ea2d4edfe75e606218114994059cee65e5f4d1b
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: MoritzLaurer <huggingface.co/MoritzLaurer>Newsgroups: alt.freeweights.visionSubject: ModernBERT-large-zeroshot-v2.0Date: 2026-11-15Commit: a51e07bSigned: 2DC28CC7B26A283B ✓
Upstream
MoritzLaurer/ModernBERT-large-zeroshot-v2.0
Pinned commit
a51e07b524299e309dd2b88d48b0cfa2bd9ec598
Manifest
ModernBERT-large-zeroshot-v2.0.json · .minisig
Manifest sha256
2574394a85d05c346c26c9e65cabbacf2ee3880e2fa258b0f64aaa0bf1555415
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 04:46 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/MoritzLaurer/ModernBERT-large-zeroshot-v2.0.json $ curl -O https://freeweights.allternit.com/manifests/MoritzLaurer/ModernBERT-large-zeroshot-v2.0.json.minisig $ minisign -Vm ModernBERT-large-zeroshot-v2.0.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

answerdotai/ModernBERT-large · this model is a quantization of it

1 model in the catalog name this one as their parent.

Quantizations
1
Quantizations 1
Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download MoritzLaurer/ModernBERT-large-zeroshot-v2.0
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. ModernBERT-large-zeroshot-v2.0.json. Provenance. Takedown history: none.
Source
huggingface.co/MoritzLaurer/ModernBERT-large-zeroshot-v2.0

Variants

1 quantized or adapter build of ModernBERT-large-zeroshot-v2.0. Each has its own license check and page.

SubjectSizeParamsLicenseDate
ModernBERT-large-zeroshot-v2.0-ONNXHosted (signed) From: onnx-community in alt.freeweights.vision4.69 GB—Permissive2026-11-15

Related models

All of alt.freeweights.vision →
From: dima806 <huggingface.co/dima806>Subject: ai_vs_real_image_detection
From: legekka <huggingface.co/legekka>Subject: AI-Anime-Image-Detector-ViT
From: onnx-community <huggingface.co/onnx-community>Subject: AI-Anime-Image-Detector-ViT-ONNX
From: AICodexLab <huggingface.co/AICodexLab>Subject: answerdotai-ModernBERT-base-ai-detector
From: onnx-community <huggingface.co/onnx-community>Subject: answerdotai-ModernBERT-base-ai-detector-ONNX
From: onnx-community <huggingface.co/onnx-community>Subject: bart-large-cnn-ONNX