Cosmos-Embed1 Collection Joint video-text embedding for physical AI ⢠5 items ⢠Updated Aug 11 ⢠9
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression Paper ⢠2609.19969 ⢠Published 5 days ago ⢠153
deepseek-ai/DeepSeek-V4.1-Flash Image-Text-to-Text ⢠763B ⢠Updated 12 days ago ⢠512k ⢠⢠3.56k
On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability Paper ⢠2608.30320 ⢠Published 22 days ago ⢠63
Qwen-Drive-1.0: An Initial Step towards a Vision-Language Foundation Model for Autonomous Driving Paper ⢠2609.00111 ⢠Published 22 days ago ⢠313
Cosmos-Reason2 Collection ā ļø This collection is archived. š https://huggingface.co/collections/nvidia/cosmos3 ⢠8 items ⢠Updated Aug 11 ⢠28
view article Article Welcome NVIDIA Cosmos 3: The First Open Omni-model for Physical AI Reasoning and Action nvidia ⢠Jun 1 ⢠90
deepseek-ai/DeepSeek-V4-Flash-Vision-Exp Image-Text-to-Text ⢠305B ⢠Updated 21 days ago ⢠803k ⢠⢠924
Cosmos3 Collection Omnimodal World Models for Physical AI ⢠6 items ⢠Updated 25 days ago ⢠202
Running Featured 84 QED-Nano: Teaching a Tiny Model to Prove Hard Theorems š 84 Who needs 1T parameters? Olympiad proofs with a 4B model