A.X K2 lands on Hugging Face with 688B total parameters and 33B active
_akhaliq · x · 2026-07-29
A.X K2 has been posted on Hugging Face as a large sparse MoE language model. The model card says it has 688B total parameters with 33B active parameters, is positioned as an agentic foundation model, and uses a Think-Fusion recipe that supports both a thinking mode and a lower-latency non-thinking mode. The card also says it is part of South Korea’s sovereign AI foundation model project.
Related event: South Korea Releases A.X K2, a 688B-Parameter Open-Source Model(2 posts)→
More from Models
- Why are sub-15B open models so rare now? Labs seem to have settled on 24-27B minimum — Hot_Example_4456 · 2026-10-16
- User spots model openly generating "ideal responses" to train other models — liminal_bardo · 2026-10-16
- Reflection's 501B MoE Beam hits 463 tok/s on 8x GB200 with NVFP4 at day 0 — zhyncs42 · 2026-10-16
- Gemma 4 31B QAT on one 24GB card: 46 tok/s plain, ~80 with draft model on RTX 4090 — Subject_Poet_3861 · 2026-10-16
- JevBench v1.7.46 leaderboard: new d1-ultra tops 32 API systems at 76.1 composite — airesearch12 · 2026-10-16
- Team of 5 mulls dropping ChatGPT Pro for Business as OpenAI quota parity stays unclear — bettercallfool · 2026-10-16