InfiX.ai

Powering the Last Mile
of Science

Our Mission and Vision

We are a research team dedicated to advancing AI through low resource training&inference ,model fusion , GUI agents, and multi modal intelligence.

Mission

At InfiX.ai, we believe in making AI more accessible, efficient, and capable. Our research focuses on developing innovative techniques that push the boundaries of what AI systems can achieve while maintaining practical applicability.

Vision

To become the global leader in high-efficiency GenAI technologies and industries.

Research Focus

Reasoning-Enhanced Low-Resource Training

We build highly capable AI systems with minimal computational cost, enabling edge deployment, local privacy-preserving processing, and lower infrastructure costs.

Model Fusion & Model Merging

We combine AI models through model merging of homogeneous architectures and model fusion across heterogeneous models to achieve higher performance and efficiency.

Latest Research

Dec 11 , 2025

From Generalist to Medical Specialist: Building Domain-Specific Multimodal LLMs via CPT, SFT and RLVR

A systematic overview of a three-stage pipeline for building medical multimodal LLMs from generalist backbones: Continual Pretraining (CPT), Supervised Fine-Tuning (SFT), and Reinforcement Learning with Verifiable Rewards (RLVR). InfiMed2-4B achieves 60.7% average accuracy across medical benchmarks, a 7.0 point improvement over the baseline.

Oct 03 , 2025

InfiAgent: A self-evolving pyramid agent framework for infinite scenarios released

With the rapid development of artificial intelligence, Large Language Model (LLM) agents have demonstrated remarkable capabilities in organizing and executing complex tasks. However, their development heavily relies on carefully designed workflows, repeatedly debugged prompts, and deep domain expertise. This highly manual approach significantly hinders the large-scale adoption and cost-effectiveness of agent-based technology across industries.

Oct 03 , 2025

Model Merging Scaling Laws: A New Way to Predict and Plan LLM Composition

We study empirical scaling laws for language model merging measured by cross-entropy. Despite wide practical use, merging has lacked a quantitative rule predicting returns as experts are added or model size scales. We identify a compact power law coupling model size and expert count: a size-dependent loss floor that decreases with capacity and an inverse‑k merging tail with clear diminishing returns. The law holds in-domain and cross-domain, tightly fits curves across architectures and methods (Average, TA, TIES, DARE), and explains two regularities: most gains arrive early and variability shrinks as more experts are included. A simple theory accounts for the ~1/k gain pattern and links floor and tail to base model properties and cross-domain diversity. This enables predictive planning: estimate experts needed for a target loss, decide when to stop, and trade off scaling the base model versus adding experts under a fixed budget—turning merging from heuristic practice into a computationally efficient, plannable alternative to multitask training. It suggests a scaling principle for distributed generative AI: predictable gains via composing specialists, offering a complementary path toward AGI-level systems.

View All Research

Co-Gen AI

Co-Gen AI reimagines how we develop, evaluate, and evolve artificial intelligence. Breaking free from centralized control, we empower communities to shape the future of AI through blockchain-powered collaboration and transparent governance.

Learn more