An instruction-tuned multi-modal model based on BLIP-2 and Vicuna-13B

• Public
• 178.9K runs
• Streaming
• GitHub
• Paper
• License