InternVL3.5 is a new family of open-source multimodal models, from 1B to a 241B flagship. Powered by Cascade Reinforcement Learning, it achieves SOTA results on reasoning and agentic tasks, with a 4x inference speedup over its predecessor.

Open MLLM family (1B-78B) from OpenGVLab. Excels at vision, reasoning, long context & agents via native multimodal pre-training. Outperforms base LLMs on text tasks.