The company said that the model was trained on 15 trillion mixed visual and text tokens.

30-person startup Arcee AI has released a 400B model called Trinity, which it says is one of the biggest open source foundation models from a US company.

Kimi has a standard mode and a Thinking mode that offers higher output quality. Additionally, a capability called K2.5 Agent ...

McGill engineering researchers have introduced an open-source model that makes it easier for experts and non-experts alike to evaluate greenhouse gas emissions from U.S. natural gas supply chains and ...

B, an open-source AI coding model trained in four days on Nvidia B200 GPUs, publishing its full reinforcement-learning stack ...

The Chinese AI start-up says its latest OCR model delivers stronger performance after adopting an Alibaba-developed ...

Kimi K2.5, an open source model with a 262k context window, helps you ship code faster with accurate refactoring and tests.

The non-profit Allen Institute for AI (AI2) has launched a family of open-source coding models targeting independent developers and SMEs.

Move over, Claude: Moonshot's new AI model lets you vibe-code from a single video upload ...

From fine-tuning open source models to building agentic frameworks on top of them, the open source world is ripe with ...

Furthermore, Nano Banana Pro still edged out GLM-Image in terms of pure aesthetics — using the OneIG benchmark, Nano Banana 2.0 is at 0.578 vs. GLM-Image at 0.528 — and indeed, as the top header ...

Moonshot AI, a Chinese large-language model developer, released its new foundation model called Kimi K2.5, saying it excels in coding as well as its understanding of visual data. K2.5, which can ...