InternVL3-78B
OpenGVLab's 78B open vision-language model with native 4K support.
About
InternVL3-78B is the flagship of InternVL3, the third-generation InternVL vision-language series from OpenGVLab, released in April 2025. It scales the family's native multimodal pretraining to 78 billion parameters with a focus on high-resolution image understanding.
InternVL3 introduced native 4K image support, improved tool-use and agentic behavior, and stronger multilingual and multi-image reasoning than InternVL2.5. The 78B tier was positioned against the strongest proprietary vision-language models on multimodal benchmarks, vendor-reported.
The InternVL family is released under MIT, which made InternVL3-78B one of the most permissively licensed large vision-language models available. It is commonly served with vLLM and LMDeploy.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 100; Trending score 78/100 from researcher assessment.