New benchmark shows top LLMs achieve only 29% pass rate on OpenTelemetry instrumentation, exposing the gap between ...

OpenAI Group PBC today introduced a platform called Frontier that companies can use to build and manage artificial ...

The company says its latest model’s agentic skills also apply to a broader set of knowledge work such as presentations and ...

On SWE-Bench Verified, the model achieved a score of 70.6%. This performance is notably competitive when placed alongside significantly larger models; it outpaces DeepSeek-V3.2, which scores 70.2%, ...

Discover Qwen3-Max Thinking, Alibaba's advanced AI model with extended reasoning capabilities. Learn about its features, ...

Today, we’re proud to introduce Maia 200, a breakthrough inference accelerator engineered to dramatically improve the economics of AI token generation. Maia 200 is an AI inference powerhouse: an ...

Kimi K2.5 adds Agent Swarm with up to 100 parallel helpers and a 256k window, so teams solve complex work faster.

Riding a motorcycle for the first time is exciting, but it can be particularly dangerous, and new riders are especially ...

Codex, a new coding model that, according to the development team, was significantly involved in its own development.

The evaluation used identical Claude Sonnet 4.5 agents under two conditions. In the baseline condition, the agent relied on native file search and tool-driven exploration to infer repository structure ...