InternVL3.5 alternatives
Models that cover what InternVL3.5 does, closest first. All of them take the same request, so trying one is a one-word change.
General vision model with strong handwriting and layout understanding at a low price.
General vision model. Good at messy scans and following extraction instructions.
General vision model with careful reading of long, dense documents.
OpenAI's flagship model (August 2025). Reads images and PDFs and follows detailed extraction instructions.
Smaller, cheaper GPT-5 model that still accepts images and PDFs.
OpenAI model from April 2025 with image input and a 1M-token context window.
Dashed tags are things InternVL3.5 does not support. Model IDs: google/gemini-flash, openai/gpt, anthropic/claude, openai/gpt-5, openai/gpt-5-mini, openai/gpt-4-1.