jevify-qwen3-vl-2b-t2 is an open-weight model from Praveen Raj U S, released under Apache License 2.0. Its published files total 69.8 MB.
A Jevify Tier 2 model: Qwen/Qwen3-VL-2B-Instruct used as a vision-language model: state may carry images (PIL, path, URL, data URI or bytes) and the same three typed questions are asked about them, with a rank-16 LoRA confined to the decoder half of the…
Model Card
By Praveen Raj U S, published under apache-2.0, revision 46e8e72c27b5.
A Jevify Tier 2 model: Qwen/Qwen3-VL-2B-Instruct used as a vision-language model: state may carry images (PIL, path, URL, data URI or bytes) and the same three typed questions are asked about them, with a rank-16 LoRA confined to the decoder half of the model, trained on the readout itself — the restricted log-softmax over the allowed answers at the answer position — with the primitive's proper scoring rule, on the aokvqa train split only. The adapter (17,432,576 parameters) is merged into the backbone at load, so it serves at the plain checkpoint's speed. The calibration recipe was fitted on validation splits of the adapted model. pope, ai2d never appeared in training. Serve it as a…
Read Praveen Raj U S's full model card
Qwen3-VL-2B, Jevified (Tier 2, decoder LoRA)
A Jevify Tier 2 model: Qwen/Qwen3-VL-2B-Instruct used as a vision-language model: state may carry images (PIL, path, URL, data URI or bytes) and the same three typed questions are asked about them, with a rank-16 LoRA confined to the decoder half of the model, trained on the readout itself — the restricted log-softmax over the allowed answers at the answer position — with the primitive's proper scoring rule, on the aokvqa train split only. The adapter (17,432,576 parameters) is merged into the backbone at load, so it serves at the plain checkpoint's speed. The calibration recipe was fitted on validation splits of the adapted model. pope, ai2d never appeared in training.
from jevify import load_jevified
model = load_jevified("Praveenrajus/jevify-qwen3-vl-2b-t2")
model.ask({"image": "https://example.com/photo.jpg", "question": "Is there a cat?"},
{"q": {"type": "noul", "instructions": "Is the answer to `question` yes, based on `image`?"}})
Serve it as a drop-in for the TypeSafe SDK: jevify-serve --model Praveenrajus/jevify-qwen3-vl-2b-t2 then TYPESAFE_BASE_URL=http://localhost:8000.
Results on jev-bench test splits
Macro accuracy 0.804, macro ECE 0.035 over 3 sources.
| source | primitive | n | acc | ECE | Brier |
|---|---|---|---|---|---|
ai2d |
choice | 1500 | 0.707 | 0.041 | 0.387 |
aokvqa |
choice | 744 | 0.813 | 0.027 | 0.270 |
pope |
noul | 2000 | 0.893 | 0.037 | 0.078 |
Recipe: mode index, permutations {'choice': 1, 'score': 1, 'noul': 1}, temperature
{'choice': 1.001, 'score': 1.0, 'noul': 1.3129}, Noul bias {'noul': 1.4}.
Predictions, metrics and figures: results/qwen3vl-2b-t2-decoder/ on
Praveenrajus/jev-bench. Findings and method:
docs/FINDINGS.md.
Identity and Version
- Repository
- Praveenrajus/jevify-qwen3-vl-2b-t2
- Publisher
- Praveen Raj U S
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- 46e8e72c27b50fe08f671bdf6b0e013acee5aae6
- First published
- 2026-09-22
- Last updated
- 2026-09-22
Files and Weights
5 files, 69.8 MB in total. The weights are 1 file totalling 69.8 MB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| lora/adapter_model.safetensors | Weights | 69.8 MB | 67c56cd13474 |
| jevify_config.json | Configuration | 1.6 KB | — |
| lora/adapter_config.json | Configuration | 1.2 KB | — |
| README.md | Documentation | 2.2 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 69.8 MB
Released by Praveen Raj U S through its official repository on Hugging Face. Read the license.
Built From
- Derived from Qwen/Qwen3-VL-2B-Instruct
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 69.8 MB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About jevify-qwen3-vl-2b-t2
Can I use jevify-qwen3-vl-2b-t2 commercially?
Yes. jevify-qwen3-vl-2b-t2 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.