Home/AI/ML Notes/GenAI and LLMs Summary/Vision & Multimodal Models
⌂ Main Page
GenAI and LLMs Summary

Vision & Multimodal Models

Image and video models, multimodal architectures, and VL-JEPA.

On this page

Image and Vision Models

Image Captioning

Text to Image

Text to Video

Best multi-modal models

1. Understanding / Analysis of multimodal content

Models for understanding multimodal content.

1.1. General-purpose proprietary models (partially fine-tunable)

1.2. General-purpose open-source models (fully fine-tunable)

The open-source gap has closed faster than expected. Qwen3-VL-235B-A22B-Instruct rivals top-tier proprietary models Gemini Pro and GPT-5 for general Q&A, 2D/3D grounding, video understanding, OCR, and document comprehension. All output is text:

1.3. CLIP-style embeddings:

2. Generation / Creation of multimodal content

Models for creation rather than analysis.

VL-JEPA

Autoregressive LLMs learn surface-level language patterns instead of pure meaning. They don’t understand meaning first — they learn to speak fluently first by just generating the next token. LLMs taught machines to speak.
VL-JEPA is teaching machines to understand. Shift from token prediction to semantic prediction.

VL-JEPA is based on the Joint Embedding Predictive Architecture (JEPA) philosophy and extends it to vision + language. Instead of predicting text tokens, it predicts meaning - semantic embeddings. VL-JEPA ensures cleaner semantic understanding - optimizes for: “What does this mean?”. Instead of talking more, models will understand continuously and respond selectively.

VL-JEPA has four key components:
1. Vision Encoder – Converts images/videos into embeddings
2. Text Encoder – Converts target text into embeddings
3. Predictor – Predicts the text embedding from vision + query
4. Lightweight Decoder – Converts embeddings to text only when needed
No text decoding is required during training.

Everything lives in a shared embedding space => VL-JEPA naturally supports:
🔹Open-vocabulary classification (no pre-defined label set)
🔹Image/video ↔︎ text retrieval
🔹Visual Q&A
🔹Discriminative reasoning
(model compares, or scores given options to decide which one best fits the input)
👉 No task-specific architectural changes required.

In experiments, VL-JEPA:
🔹Uses ~50% fewer parameters
🔹Matches or outperforms traditional VLMs
🔹Beats CLIP-style models in retrieval & classification
🔹Competes strongly on VQA tasks

Lower latency, reduced compute cost since VL-JEPA doesn’t decode tokens continuously and decodes text only on demand. It enables continuous real-time understanding and is perfect for: a) Video understanding, b) Robotics, c) AR/VR, d) Always-on AI agents.

VL-JEPA aligns perfectly with Agentic AI, edge & on-device AI, real-time multimodal reasoning