MBZUAI releases K2 Horizon, a six-model fleet with the 0.9B scoring over 48 on AIME 2026

K2 Horizon: Frontier Performance, Radically Open

IFM at MBZUAI released K2 Horizon, a six-model fleet from 0.9B to 375B-A23B. The 0.9B, 3.7B, and 7B models set new SOTA in their size classes; the 0.9B scored above 48 on AIME 2026 with reasoning and tool-use capabilities. The 36B-A4B uses a new MoVA attention mechanism, outperforming larger models per active parameter. This is a full open-science release: intermediate checkpoints, data recipes, code, logs, and evals from pretraining through agentic post-training, under Apache 2.0. The post doesn't disclose specific benchmark comparison numbers or latency data, so real-world performance still needs third-party validation.

Why it matters: IFM dropped six fully open models at once, with the 0.9B hitting 48+ on AIME 2026 math and the 36B introducing a new MoVA attention mechanism — high information density. Not scoring 85+ because IFM isn't an OpenAI/Anthropic-tier lab yet and market validation hasn't caught up; ...