# Vev 4B v2 decision model by Wang Xiaolong > Vev 4B v2 returns decision probabilities for text and images. Check its noncommercial weights, custom attention and reported calibration limits. Updated 2026-11-07 (Australia/Adelaide) ## Quick answers ### What is Vev 4B v2 by Wang Xiaolong? [Vev 4B v2](): A Qwen3.5-4B text/image decision fine-tune with merged weights and a Jev-like API. [Vev 4B v2 facts]() [Source]() ### Who makes Vev 4B v2? [Vev 4B v2](): Wang Xiaolong [Vev 4B v2 facts]() [Source]() ### Are Vev 4B v2's weights open? [Vev 4B v2](): Public merged weights; no separate adapter Licence: CC BY-NC 4.0; Apache-2.0 base licence retained [Vev 4B v2 facts]() [Source]() By Wang Xiaolong · Weights upload: 2026-11-06 · NON-COMMERCIAL A Qwen3.5-4B text/image decision fine-tune with merged weights and a Jev-like API. Its runtime uses bidirectional attention within the state and repeats some inputs. Plain causal inference produces different probabilities. This research preview is noncommercial. ## Facts - Size: 4B; merged weights - Inputs: English or Chinese text, JSON and images - Runtime: vev serve on an NVIDIA GPU; checkpoint-specific attention and repetition settings - Licence: CC BY-NC 4.0; Apache-2.0 base licence retained - Access: Public merged weights; no separate adapter ## Claims and sources - **VENDOR-REPORTED**: The maker reports public vision skill 72.90 and text skill 60.32, with disclosed training/evaluation overlaps and a text-suite hash mismatch outside the index. Across 18 release sets, ECE ranges from 0.01 to 0.24; four sets regress versus Vev 4B. No full private-board score is supplied. [source]() ## Links - [Source: Model card and weights]() ## Related projects No related projects are listed in the registry.