Kimi K2.7-Code claims 30% fewer thinking tokens and a drop-in API swap path, but independent benchmarks show kernel ...

Anthropic’s Fable and Mythos models have been withdrawn following US export controls, but OpenRouter might have a solution. The company has launched ...

First comparative benchmark of AI agents for API bug detection shows strong performance on simple checks, but major gaps on cross-field and business-logic failures SAN FRANCISCO, June 3, 2026 ...

M3 demonstrates that the next phase of agent development will not just be driven by larger datasets, but by efficient ...

Fusion uses multiple AI models together to improve output quality at lower cost. OpenRouter debuts Fusion, a multi-model composite AI Fusion outperforms single models, scoring 69% on DRACO benchmark ...

As data sources and volumes grow, and as a data-driven orientation is increasingly deemed to be a competitive necessity, the war between platform vendors to provide the primary repository for our data ...

Today marks an exciting moment for the developer community as xAI officially introduces the Grok Voice Agent API, opening the door for anyone to build powerful, real-time voice agents with ease.

Use left and right arrow keys to seek audio. UL has just announced that its 3DMark benchmark suite now runs on macOS, and while it was already available on iOS, the company saw that many users were ...

Developers and ChatGPT subscribers who relied on GPT-5.2 Thinking now face a forced migration. OpenAI has replaced GPT-5.2 Thinking with GPT-5.4 Thinking for Plus, Team, and Pro users, while ...