This model is a conversion of MoritzLaurer/roberta-base-zeroshot-v2.0-c to ONNX format using the Optimum library.
von is an open-weight model for zero-shot classification from Verdict, released under Apache License 2.0. Its published files total 1.6 GB.
The Von System One decision model (ModernBERT-large, 395M, English, 8192-token context), exported to ONNX fp32 and packaged for Verdict -- a high-performance serving engine for System One decision models.
Parameters—
Context—
Weights3.0 MB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads—
Model Card
By Verdict, published under apache-2.0, revision 658d0fb2ab78.
The Von System One decision model (ModernBERT-large, 395M, English, 8192-token context), exported to ONNX fp32 and packaged for Verdict -- a high-performance serving engine for System One decision models. Serve it with one command: Pick the image that matches the host. The -v mounts cache the model, so only the first run downloads it. Add -e VERDICTUSEMODELSCOPE=1 and use sgyaqing/von to fetch from ModelScope instead of the hub; add -e HFENDPOINT=https://hf-mirror.com if the hub is slow or unreachable. Endpoints: the standard decision protocols POST /v1/systemone and POST /v1/decisions, native POST /v1/choice | /v1/score | /v1/noul, plus an OpenAI-compatible layer (/v1/chat/completions…
Read Verdict's full model card
Von (Verdict-ready)
The Von System One decision model (ModernBERT-large, 395M, English, 8192-token context), exported to ONNX fp32 and packaged for Verdict -- a high-performance serving engine for System One decision models. Serve it with one command:
Pick the image that matches the host. The -v mounts cache the model, so only the first
run downloads it.
# CPU host
docker run --rm -p 8080:8080 \
-v ~/.cache/verdict:/root/.cache/verdict \
sgyaqing/verdict:0.2.0-cpu serve verdict-models/von --host 0.0.0.0
# GPU host
docker run --rm --gpus all -p 8080:8080 \
-v ~/.cache/verdict:/root/.cache/verdict \
sgyaqing/verdict:0.2.0-cuda serve verdict-models/von --host 0.0.0.0
Add -e VERDICT_USE_MODELSCOPE=1 and use sgyaqing/von to fetch from ModelScope instead of the
hub; add -e HF_ENDPOINT=https://hf-mirror.com if the hub is slow or unreachable.
Endpoints: the standard decision protocols POST /v1/systemone and POST /v1/decisions, native POST /v1/choice | /v1/score | /v1/noul, plus an OpenAI-compatible layer (/v1/chat/completions, /v1/models).
Performance
Measured on one machine (RTX 4090, AMD EPYC 7543, 16 threads), end to end over HTTP with
oha and an identical request body for both servers. The
baseline is the model author's own server (von serve, von-sdk, --no-chains); the Verdict
column is this package. Latencies are per-request P50 / P99 at the stated concurrency. The
ratios are oriented so that above 1× means Verdict is ahead in both: throughput ratio =
Verdict ÷ baseline, latency ratio = baseline ÷ Verdict.
| concurrency | Verdict rps | Verdict P50 / P99 | von rps |
von P50 / P99 |
ratio rps | ratio P50 / P99 |
|---|---|---|---|---|---|---|
| 1 | 113.8 | 8.61 / 11.98 ms | 46.4 | 22.89 / 28.46 ms | 2.45× | 2.66 / 2.38× |
| 8 | 536.4 | 14.86 / 16.61 ms | 31.2 | 250.91 / 435.99 ms | 17.2× | 16.9 / 26.2× |
| 64 | 1476.1 | 42.58 / 53.24 ms | 22.3 | 2848.80 / 3333.40 ms | 66.1× | 66.9 / 62.6× |
Files
| File | Purpose |
|---|---|
model.json |
manifest: graph file name, budgets, family recipe (independent_options), calibration map |
von.onnx + von.onnx.data |
the ONNX graph (fp32, external weights) |
tokenizer/tokenizer.json (+ tokenizer_config.json) |
the tokenizer |
SHA256SUMS |
checksums for every file above |
Provenance & fidelity
Weights are unchanged from wfzyx/von; this repo adds only the ONNX export and the serving manifest. Conversion and serving are verified against the official model (von-sdk 1.3.7): identical answers on CPU (29-case golden corpus, logit parity ≤ 4.6e-5) and on CUDA (RTX 4090, logit parity ≤ 7.4e-5, batch invariance bitwise).
License
Apache-2.0, inherited from the base model. Credit for the model itself belongs to the original author; this repository only repackages it for serving.
Identity and Version
- Repository
- verdict-models/von
- Publisher
- Verdict
- Task
- Zero-shot classification
- Modality
- Text
- Library
- Not stated by the source
- Parameters
- Not stated by the source
- Languages
- von
- Revision
- 658d0fb2ab7863eee5f2fa090be4edca43d96f86
- First published
- 2026-10-17
- Last updated
- 2026-10-17
Files and Weights
9 files, 1.6 GB in total. The weights are 1 file totalling 3.0 MB in onnx.
Weights1 file · 3.0 MB
Configuration1 file · 464 B
Tokenizer2 files · 3.6 MB
Documentation1 file · 3.2 KB
Other2 files · 1.6 GB
Repository2 files · 2.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| von.onnx | Weights | 3.0 MB | daa80b072019 |
| model.json | Configuration | 464 B | — |
| README.md | Documentation | 3.2 KB | — |
| SHA256SUMS | Other | 497 B | — |
| von.onnx.data | Other | 1.6 GB | d5b804c5ceb8 |
| .gitattributes | Repository | 1.6 KB | — |
| .ms_upload_cache | Repository | 1.1 KB | — |
| tokenizer/tokenizer.json | Tokenizer | 3.6 MB | — |
| tokenizer/tokenizer_config.json | Tokenizer | 571 B | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 3.0 MB
Download from Verdict
Released by Verdict through its official repository on Hugging Face. Read the license.
Built From
- Derived from wfzyx/von
- Quantized from wfzyx/von
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 3.0 MB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About von
Can I use von commercially?
Yes. von is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
Similar Models
https://huggingface.co/facebook/bart-large-mnli with ONNX weights to be compatible with Transformers.js. If you haven't already, you can install the Transformers.js JavaScript library from NPM using: Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).
distilbart-mnli is the distilled version of bart-large-mnli created using the No Teacher Distillation technique proposed for BART summarisation by Huggingface, here. We just copy alternating layers from bart-large-mnli and finetune more on the same data. This is a very simple and effective technique, as we can see the performance drop is very little. Detailed performace trade-offs will be posted in this sheet. If you want to train these models yourself, clone the distillbart-mnli repo and follow the steps below Clone and install transformers from source Download MNLI data Create student model Start fine-tuning You can find the logs of these trained models in this wandb project.
distilbart-mnli is the distilled version of bart-large-mnli created using the No Teacher Distillation technique proposed for BART summarisation by Huggingface, here. We just copy alternating layers from bart-large-mnli and finetune more on the same data. This is a very simple and effective technique, as we can see the performance drop is very little. Detailed performace trade-offs will be posted in this sheet. If you want to train these models yourself, clone the distillbart-mnli repo and follow the steps below Clone and install transformers from source Download MNLI data Create student model Start fine-tuning You can find the logs of these trained models in this wandb project.
https://huggingface.co/typeform/mobilebert-uncased-mnli with ONNX weights to be compatible with Transformers.js. If you haven't already, you can install the Transformers.js JavaScript library from NPM using: Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).
https://huggingface.co/cross-encoder/nli-deberta-v3-xsmall with ONNX weights to be compatible with Transformers.js. If you haven't already, you can install the Transformers.js JavaScript library from NPM using: Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).