Blog

01.AI's Yi 34B model delivers exceptional bilingual performance, competing directly with larger models like Llama 2 70B while maintaining cost efficiency.

Zhipu AI releases ChatGLM3-6B, a groundbreaking open-source model featuring function calling, code interpretation, and agent capabilities.

Discover how HuggingFace's Zephyr 7B leverages Direct Preference Optimization to deliver state-of-the-art performance in a compact 7B parameter package.

Discover how Mistral AI's 7B model achieved remarkable performance gains over much larger competitors, revolutionizing the open-source AI landscape.

Alibaba's Qwen 72B emerges as a formidable open-source competitor, delivering exceptional performance across Chinese and English tasks while offering true open weights access.

WizardCoder emerges as the top open-source coding model, achieving remarkable HumanEval scores that outperform commercial alternatives through innovative Evol-Instruct tuning.

Meta AI releases Code Llama 34B, a specialized Llama 2 variant designed for code generation with 100K context window support across multiple programming languages.

Meta's Llama 2 became the first truly open-weight large model for commercial use, launching the modern open LLM ecosystem with 7B, 13B, and 70B parameter variants.

Discover Claude 2's revolutionary 200K context window and Constitutional AI approach that's reshaping enterprise AI applications.