K2 Horizon ships six fully open models from 0.9B to 375B, data and recipes included
HongyiWang10 · x · 2026-09-04
AI startup IFM announced K2 Horizon, billed as the largest fully open-source model launch in AI history: six foundation models ranging from 0.9B dense to 375B MoE. On coding and agentic tasks the fleet claims top-tier performance per size class, with the 0.9B, 3.7B and 7B models setting new SOTA at their scales. The team stresses radical openness—code, training data and recipes are all public—aiming for frontier-level performance while supporting open science. Launch page, tech blog and Hugging Face weights are live.
Related event: MBZUAI's IFM Releases K2 Horizon: Six Fully Open Models from 0.9B to 375B(30 posts)→
More from Models
- Users want open-format export from Claude Docs as charts only save as images — BlackHC · 2026-10-20
- EMNLP study: LLM safety interventions drop up to 4.5x when distress is wrapped in delusion — MaartenSap · 2026-10-20
- Cybersecurity-focused VLM OrcaCyber-Nano-1.0 trends on Hugging Face — orcarouter · 2026-10-20
- 0.8B local model with 2,000 labels beats Jev's frontier API on banking data — kalyan_kpl · 2026-10-20
- New TTS benchmark tests 26 models across 30,564 requests: fastest at 46ms — mattturck · 2026-10-20
- Agents handle unreleased language Jai flawlessly, even compile-time execution — lucasmeijer · 2026-10-20