-
Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception
Paper ⢠2602.11858 ⢠Published ⢠62 -
inclusionAI/ZwZ-4B
Image-Text-to-Text ⢠5B ⢠Updated ⢠292 ⢠34 -
inclusionAI/ZwZ-8B
Image-Text-to-Text ⢠9B ⢠Updated ⢠189 ⢠50 -
inclusionAI/ZwZ-RL-VQA
Viewer ⢠Updated ⢠111k ⢠1.73k ⢠17
Collections
Discover the best community collections!
Collections trending this week
-
MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases
Paper ⢠2402.14905 ⢠Published ⢠134 -
facebook/MobileLLM-125M
Text Generation ⢠Updated ⢠2.34k ⢠138 -
facebook/MobileLLM-350M
Text Generation ⢠Updated ⢠409 ⢠37 -
facebook/MobileLLM-600M
Text Generation ⢠Updated ⢠26 ⢠30
-
LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding
Paper ⢠2404.16710 ⢠Published ⢠80 -
facebook/layerskip-llama2-7B
Text Generation ⢠7B ⢠Updated ⢠322 ⢠18 -
facebook/layerskip-llama2-13B
Text Generation ⢠13B ⢠Updated ⢠54 ⢠6 -
facebook/layerskip-llama2-70B
Text Generation ⢠69B ⢠Updated ⢠29 ⢠5
-
Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception
Paper ⢠2602.11858 ⢠Published ⢠62 -
inclusionAI/ZwZ-4B
Image-Text-to-Text ⢠5B ⢠Updated ⢠292 ⢠34 -
inclusionAI/ZwZ-8B
Image-Text-to-Text ⢠9B ⢠Updated ⢠189 ⢠50 -
inclusionAI/ZwZ-RL-VQA
Viewer ⢠Updated ⢠111k ⢠1.73k ⢠17
-
MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases
Paper ⢠2402.14905 ⢠Published ⢠134 -
facebook/MobileLLM-125M
Text Generation ⢠Updated ⢠2.34k ⢠138 -
facebook/MobileLLM-350M
Text Generation ⢠Updated ⢠409 ⢠37 -
facebook/MobileLLM-600M
Text Generation ⢠Updated ⢠26 ⢠30
-
LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding
Paper ⢠2404.16710 ⢠Published ⢠80 -
facebook/layerskip-llama2-7B
Text Generation ⢠7B ⢠Updated ⢠322 ⢠18 -
facebook/layerskip-llama2-13B
Text Generation ⢠13B ⢠Updated ⢠54 ⢠6 -
facebook/layerskip-llama2-70B
Text Generation ⢠69B ⢠Updated ⢠29 ⢠5