A research team at Berkeley has introduced an innovative artificial intelligence model, DeepScaler, that challenges ...

The starting point of the project was Qwen2.5-32B-Instruct, an open-source LLM released by Alibaba Group Holding Ltd. last year. The researchers created s1-32B by customizing Qwen2.5-32B-Instruct ...

While Hugging Face cloned OpenAI's Deep Research in 24 hours, a multi-institutional team of researchers built an o1 ...

Researchers developed the S1 reasoning AI using less than $50 in compute cost to achieve a reasoning model as powerful as ...

Researchers at Stanford and the University of Washington have developed a model that performs comparably to OpenAI o1 and ...

According to OpenAI’s tests, the new version of o1 reduces “major errors” on “difficult real-world questions” by 34% compared to its preview version. If you’ve got reams of highly ...

OpenAI is restructuring its AI strategy to focus solely on GPT-5, consolidating capabilities like reasoning, voice synthesis, ...

With a few hundred well-curated examples, an LLM can be trained for complex reasoning tasks that previously required thousands of instances.

The S1-32B model represents a paradigm shift in AI reasoning capabilities, introducing an innovative technique called ...

In response to pressure from rivals including Chinese AI company DeepSeek, OpenAI is changing the way its newest AI model, o3 ...

Everyone can now use ChatGPT search without logging in - you don't even need an account with OpenAI to use its AI search ...

OpenAI has expanded o3-mini’s reasoning display in ChatGPT, revealing more of its decision-making while keeping key details ...