aiqbench takeAI Products

IFM's release of the K2 Horizon suite marks a significant shift in open-source strategy by providing a unified architecture across six model sizes, from 0.9B to 375B parameters. The inclusion of extensive training logs, intermediate checkpoints, and synthetic reasoning data offers researchers unprecedented transparency into modern scaling laws and architectural innovations like Mixture-of-Value Attention. This comprehensive package enables seamless prototyping and scaling without altering serving stacks, addressing a critical bottleneck in current open-weight model deployment.

Sources coveringMarkTechPost, Latent Space, AWS

What this isA model's reading of the article and related coverage, written for aiqbench. Not the publisher's words.

IFM Releases K2 Horizon: Six Apache 2.0 Models From 0.9B to 375B

MarkTechPostSeptember 7, 20264 min read

Most open model launches release one checkpoint and a benchmark table. The Institute of Foundation Models (IFM) released something wider last week. IFM is the frontier lab launched by MBZUAI in May 2025. K2 Horizon is a fleet of six models: 375B-A23B, 36B-A4B, 32B, 7B, 3.7B and 0.9B. Shipping alongside them are the pre-training corpus, […] The post IFM Releases K2 Horizon: Six Apache 2.0 Models From 0.9B to 375B appeared first on MarkTechPost.

Most open model launches release one checkpoint and a benchmark table. The Institute of Foundation Models (IFM) released something wider last week. IFM is the frontier lab launched by MBZUAI in May 2025. K2 Horizon is a fleet of six models: 375B-A23B, 36B-A4B, 32B, 7B, 3.7B and 0.9B. Shipping alongside them are the pre-training corpus, intermediate checkpoints, training code, configs and fine-grained logs. IFM calls it the largest fully open-source model…

Excerpt shown under the publisher's RSS terms · 71 words · full text and figures are on the publisher's site

Read at MarkTechPost ↗Ask about this article