← all releases
Ant Group's AGI initiative·Feb 1, 2026·v2.5-1T·open source
Ling-2.5-1T
Ling-2.5-1T is an advanced language model developed by Ant Group's Artificial General Intelligence (AGI) initiative, inclusionAI. It represents a significant advancement in AI, offering enhanced efficiency and performance for a wide range of applications.
Agentic AIMLLLMs
overview
Ling-2.5-1T is an advanced language model developed by Ant Group's Artificial General Intelligence (AGI) initiative, inclusionAI. It represents a significant advancement in AI, offering enhanced efficiency and performance for a wide range of applications.
key features
- 011 trillion total parameters with 63 billion active parameters
- 02Hybrid Linear Attention architecture for improved throughput
- 03Supports context lengths up to 1 million tokens
- 04Enhanced reasoning capabilities surpassing previous models
- 05Refined alignment strategies for better instruction following
related productsbrowse all →
AIBuildAIopen sourceAIBuildAI is an AI agent that autonomously constructs AI models. Given a specific task, it initiates an agent loop to analyze the problem, design models, and execute training processes, all without human intervention.Step-Audio-R1open sourceStep-Audio-R1 is an advanced audio language model developed by StepFun AI, designed to enhance audio reasoning capabilities by grounding its reasoning in acoustic features. It introduces Modality-Grounded Reasoning Distillation (MGRD), an iterative training framework that shifts the model's reasoning from textual abstractions to acoustic properties, effectively addressing the 'inverted scaling' problem where performance degrades with longer reasoning. This model has demonstrated superior performance across various audio understanding and reasoning benchmarks, surpassing models like Gemini 2.5 Pro and achieving results comparable to Gemini 3 Pro.SINQopen sourceSINQ (Sinkhorn-Normalized Quantization) is a novel, fast, and high-quality quantization method designed to make any Large Language Model (LLM) smaller while preserving accuracy. It offers a plug-and-play, model-agnostic technique that delivers state-of-the-art performance for LLMs without sacrificing accuracy.