vlmhub

vlmhub provides a unified Python interface for running multimodal prompts (interleaved text and images) on various vision-language model backends, both local and cloud-based. It supports multiple providers like Ollama, MLX-VLM, vLLM, HuggingFace Transformers, Gemini, Vertex AI, OpenAI, and Anthropic through a consistent API, enabling easy switching and integration without rewriting inference code.