AIOZ Network Report: July 2026
July continued the expansion of AIOZ AI with new datasets, models, and community challenges. These additions strengthen the resources available for building, evaluating, and deploying AI across a growing range of languages, modalities, and application domains.
This report highlights the key updates from July.
AIOZ AI Datasets
July introduced new datasets spanning long-context evaluation, multilingual language understanding, spoken intent recognition, and comprehensive language model benchmarking.
KLUE expands multilingual resources with a benchmark suite for Korean natural language understanding.
LongBench provides a benchmark for evaluating long-context language models across a diverse set of reasoning and comprehension tasks.
MInDS-14 is a multilingual spoken intent dataset designed for training and evaluating intent classification systems across multiple languages.
BIG-bench brings one of the most comprehensive language model evaluation suites to AIOZ AI, featuring 204 benchmark tasks that span reasoning, knowledge, language understanding, and emerging capabilities.
These additions continue to expand the range of datasets available on AIOZ AI, supporting model development, benchmarking, and multilingual AI research.
New Models on AIOZ AI
July introduced new models across software engineering, multimodal AI, multilingual reasoning, and conversational intelligence.
Qwen2.5-Omni is a multimodal model capable of understanding and generating content across text, images, audio, and video.
Devstral Small 2 is an agentic coding model designed for complex software engineering workflows.
SmolLM3 is a compact multilingual language model designed for long-context reasoning. Combining efficient deployment with multilingual understanding and extended context processing, it supports a wide range of practical language AI workflows.
Olmo 3.1 32B Instruct expands the AIOZ AI model library with an open model optimized for fast, agentic dialogue.
These new models broaden the range of models available on AIOZ AI, spanning software engineering, multimodal intelligence, multilingual reasoning, and conversational AI.
AIOZ AI Challenge
The Smoker Classification Challenge is live on AIOZ AI.
The challenge focuses on binary computer vision classification, where participants develop models to distinguish smoker and non-smoker images. It provides practical experience with image preprocessing, supervised learning, and visual classification workflows using a balanced dataset.
Join the challenge:
What's Next
July expanded the AIOZ AI ecosystem through new datasets, models, and community challenges, providing builders with more resources to develop, evaluate, and deploy AI applications. As the platform continues to grow, AIOZ AI remains focused on supporting practical AI workflows across research, development, and real-world deployment.
Explore more and start building on AIOZ.
About the AIOZ Network
AIOZ Network is a DePIN for AIOZ AI, AIOZ Storage, AIOZ Pin and AIOZ Stream.
Powered by a global community of AIOZ DePINs, AIOZ rewards you for sharing your computational resources for storing, transcoding, and streaming digital media content and powering decentralized AI computation.
Find Us
AIOZ All Links | Website | X | Telegram