Multimodal

Page 1 of 3

  1. DeepSeek Vision API: The $0.22/M-Token Vision Model That Uses 10x Less KV Cache

    DeepSeek's deepseek-v4-flash-vision-exp brings vision to the V4-Flash line: a 284B-parameter MoE model (13B active) that bills images at just $0.22 per million input …

    AI

  2. Seedance 2.5: ByteDance's 30-Second AI Video Model That Accepts 50 References

    Seedance 2.5 generates native 30-second video in one pass, accepts up to 50 multimodal references, and adds region-level editing. Here's how it compares to Veo 3.1 and …

    AI

  3. VILA: NVIDIA's Open-Source Vision Language Model Family from NVlabs

    Vision Language Models (VLMs) that can reason about both images and text have become one of the most active areas in AI research. VILA (Visual...

    AI

  4. SGLang Omni: Multimodal LLM Inference with SGLang

    Multimodal AI — models that understand images, audio, and video alongside text — has moved from research novelty to production necessity. Document...

    AI

  5. Seed1.5-VL: ByteDance's Vision-Language Foundation Model Achieving 38 SOTA Benchmarks

    In the rapidly advancing field of vision-language models, a new heavyweight has emerged from an unexpected corner. Seed1.5-VL, developed by...

    AI

  6. Qwen2.5-Omni: Alibaba's End-to-End Multimodal AI Model

    Qwen2.5-Omni is Alibaba's flagship open-source multimodal AI model, developed by the QwenLM team at Alibaba Cloud. As a single end-to-end model...

    AI