Model Gallery

Discover and install AI models from our curated collection

2 models available
1 repositories
Documentation

Find Your Perfect Model

Filter by Model Type

Browse by Tags

runtimelabs-kayak-qwen3-vl-4b-q4
RuntimeLabs Kayak is a 4B Qwen3-VL fine-tune for screenshot grounding on Kayak flight-search interfaces. It predicts a single action and a normalized bounding box from an image and instruction. This Q4_K_M GGUF build includes the F16 vision projector and uses an 8,192-token context with llama.cpp.

Repository: localaiLicense: apache-2.0

runtimelabs-kayak-qwen3-vl-4b-q8
RuntimeLabs Kayak is a 4B Qwen3-VL fine-tune for screenshot grounding on Kayak flight-search interfaces. It predicts a single action and a normalized bounding box from an image and instruction. This Q8_0 GGUF build includes the F16 vision projector and uses an 8,192-token context with llama.cpp.

Repository: localaiLicense: apache-2.0