K2 Horizon
### TL;DR
K2 Horizon is a connected fleet of six open-source AI models ranging from 0.9B to 375B parameters, designed for tasks spanning edge devices to enterprise-grade reasoning. The release is uniquely transparent, providing full access to training data, intermediate checkpoints, training logs, and methodologies to support open science and reproducibility.
Key Insights & Metrics
Key Features
- Fully open-source release including weights, training data, code, and intermediate checkpoints
- Mixture-of-Value-Attention (MoVA) architecture for high-efficiency sparse performance
- Diffusion distillation for lossless 3x inference speedup
- Unified architecture and interface across all six model sizes for seamless scaling
- State-of-the-art performance in small-scale classes (0.9B, 3.7B, 7B)
- Comprehensive support for agentic workflows and tool-use
Master AI Marketing: Work Smarter, Not Harder
Unlock the power of AI to automate workflows and scale your results. Join 1.5M+ professionals and stay ahead of the curve with the latest AI insights.
→ Related Releases
MiroThinker
MiroThinker is an open-source search agent model developed by MiroMindAI, designed for tool-augmented reasoning and real-world information seeking. It aims to match the deep research capabilities of leading AI models like OpenAI's Deep Research and Google's Gemini Deep Research.
Letta Code SDK
The Letta Code SDK is a software development kit that enables developers to build deeply personalized agents with persistent memory that learn over time. It serves as the interface to Letta Code, facilitating the creation of stateful agents capable of continuous learning and improvement.
OB-1
OB-1 is a self-improving coding agent developed by OpenBlock Labs, designed to autonomously handle the full development lifecycle, from project management to pull requests. It integrates seamlessly into existing workflows, enhancing productivity and code quality.
AIO Sandbox
AIO Sandbox is an integrated environment designed for AI agents, combining a browser, terminal, filesystem, VSCode, Jupyter, and MCP Server into a single Docker container. This unified setup allows seamless development and execution of AI agents without the need for multiple services.
Discussion
Sign in to leave a review
Reviews
No reviews yet. Be the first to review!