Doubao 1.5 Vision Pro vs Skywork-R1V3-38B

ByteDance

Doubao 1.5 Vision Pro, released by ByteDance via Volcano Engine on January 22, 2025, is a multimodal large language model from the Doubao 1.5 family with comprehensive upgrades to visual reasoning, OCR, and fine-grained image understanding. Built on a large-scale sparse MoE architecture, it targets vision-intensive workflows including document analysis, chart interpretation, and visual question answering.

Kunlun Tech

Skywork-R1V3-38B, released by Kunlun Tech's Skywork AI team on July 9, 2025, is a 38 billion parameter multimodal reasoning model built on InternVL-38B with reinforcement learning post-training that enhances both visual and textual reasoning. It uses the GRPO algorithm and cold-start fine-tuning to improve reasoning across image and text modalities. Skywork-R1V3-38B targets open-source multimodal reasoning deployments requiring strong performance across vision-language benchmarks.