Korean Company Motif Releases 314B MoE Open-Weight Model, Rivals DeepSeek V4 Pro
eliebakouch · x · 2026-07-21
Korean company Motif releases Motif 2.6B tech report, a 13B active 314B total MoE model, performing on par with Minimax M3 and DeepSeek V4 Pro. It features differential attention (GDLA), per-expert activation function polynorm, modified mHC. Trained on 2.5T tokens with data mixture scheduling and WSD. Fine-tuning uses EvolKit and dataset fusion.
Related event: Korea's Motif Releases Motif-3 Beta with 314B Parameters on Hugging Face(4 posts)→
More from Models
- Anthropic's Terra returns with better writing and 2X usage vs 6.1 Sol — weswinder · 2026-10-27
- User catches model failing to parse 400-paper file, citing stale news sources instead — jungofthewon · 2026-10-27
- Odd eval result: Gemini 4.1 Flash gets more accurate when asked to control its CoT — dfrsrchtwts · 2026-10-27
- Poster surprised that v4.1 flash gets more accurate when asked to control its CoT — dfrsrchtwts · 2026-10-27
- Dev claims Anthropic is baking 'sabotage' into model behavior for 'wrong' engineering — cephaloform · 2026-10-27
- Gemini fails at listing unread emails, tells user to search Gmail manually — gethackteam · 2026-10-27