Technology ❯Artificial Intelligence ❯Benchmarking

Performance Metrics

MMLU and HumanEval Scores ARC-AGI Image Generation Models

Alibaba Unveils Qwen2.5-Omni-7B AI Model for Multimodal Use on Everyday Devices

The open-source model processes text, images, audio, and video, featuring real-time speech and video chat capabilities with innovative 'Thinker-Talker' architecture.

6 ARTICLES

5mo ago

Meta Releases Llama 3.3, a Smaller, Cost-Efficient AI Model That Matches Top Competitors