Deriva Series The Deriva Series of models is our most advanced open source AI series. Deriva is our vision capable general model with Image, Video, Text input. Qwen/Qwen3.8-27B Image-Text-to-Text • 28B • Updated Aug 14 • 7.15M • • 16k
Hikaru Series The Hikaru Series of models are built using our custom ERL (Explanatory Reinforcement Learning). Hikaru-1 is built off Ministral-3 (8b). mistralai/Ministral-3-8B-Instruct-2512 9B • Updated Jul 15 • 70.1k • 204
Deriva Series The Deriva Series of models is our most advanced open source AI series. Deriva is our vision capable general model with Image, Video, Text input. Qwen/Qwen3.8-27B Image-Text-to-Text • 28B • Updated Aug 14 • 7.15M • • 16k
Hikaru Series The Hikaru Series of models are built using our custom ERL (Explanatory Reinforcement Learning). Hikaru-1 is built off Ministral-3 (8b). mistralai/Ministral-3-8B-Instruct-2512 9B • Updated Jul 15 • 70.1k • 204
