on-device open-weight LLM (Apache 2.0) with 1.5B effective and 3.1B total parameters
AI & ML interests
None defined yet.
Recent Activity
Papers
A2Z GameSpec-Bench: How Faithfully Can Coding Agents Generate Games from Game Design Specifications?
WHALE: A Simple Recipe for Joint Harness-Weight Optimization
a 30B-A4B open-weight LLM (Apache 2.0) with an MoE + linear-attention architecture.
-
KRAFTON/Raon-1-mini-preview
Text Generation • 29B • Updated • 5.13k • 248 -
KRAFTON/Raon-1-mini-preview-gguf
Text Generation • 29B • Updated • 48.6k • 85 -
KRAFTON/Raon-1-mini-preview-mlx
Text Generation • 29B • Updated • 2.89k • 31 -
KRAFTON/Raon-1-mini-preview-nvfp4
Text Generation • 15B • Updated • 1.64k • 26
on-device open-weight LLM (Apache 2.0) with 1.5B effective and 3.1B total parameters
a 30B-A4B open-weight LLM (Apache 2.0) with an MoE + linear-attention architecture.
-
KRAFTON/Raon-1-mini-preview
Text Generation • 29B • Updated • 5.13k • 248 -
KRAFTON/Raon-1-mini-preview-gguf
Text Generation • 29B • Updated • 48.6k • 85 -
KRAFTON/Raon-1-mini-preview-mlx
Text Generation • 29B • Updated • 2.89k • 31 -
KRAFTON/Raon-1-mini-preview-nvfp4
Text Generation • 15B • Updated • 1.64k • 26
a bilingual English/Korean speech language model with approximately 21.2B total parameters and 3.5B active parameters