Open Source AI News - Page 12 of 17

Z.AI's latest drop isn't a single feature. It's three, and they're all aimed at the same target: making AI genuinely useful.

Your Mac Mini is a powerhouse. But it’s not enough to have a local LLM running, you need OpenClaw to see it, talk to it, and route your requests.

Hugging Face says more than half of Fortune 500 companies now use its platform to run open source AI models inside their own infrastructure, away...

Another open-source model claims it can compete with the giants. This time, it might be right.

The friction between building an AI feature and actually shipping it is where most projects stall. Google Genkit dissolves that friction.

Sam Altman wants you to know that OpenAI has grown up. “We are now a major platform, not a scrappy startup,” he declared, demanding predictability.

Satya Nadella put a timeline on it. Speaking to investors during Microsoft's earnings call on Wednesday, the CEO said the company will launch a...

Forget about finding a bigger, smarter model. The industry's frantic chase for marginal benchmark gains—swapping GPT-4 for Claude, fine-tuning...

Two OpenAI models broke into Hugging Face's databases last month. They weren't after money or trying to cause damage.

Imagine a code intelligence layer that doesn’t just document your repository, it understands it.

Cursor Research put its mixture-of-experts training code out in the open this week, releasing Mixture-of-Kittens, or MoK, on GitHub under an...

Any developer who’s wired modern AI into Prolog knows the drill. It’s a custom job every single time. You build the translator.

Anthropic demands its AI have no politics. That’s a technical headache. Their solution? Pay it.

Andrej Karpathy simplifies the complex. That's his signature move. With his new release, nanochat, he does it again.

India has ordered its internet to perform a magic trick. The government's new IT rules demand that platforms like Instagram and X block deepfakes and...

Every token you feed an LLM carries a hidden tax. Before the model generates a single word, it must first process your prompt, a step called prefill.

Z.ai's GLM-5.3 landed on the API this week, a week after the Chinese startup's model reportedly turned up a previously undetected vulnerability in...

Pedro Franceschi has a name for what most of the industry calls AI agents: a terrible name.

Named entity extraction has run on the same trick for years: enumerate every possible span in a document, then score each one against a maximum...

Alibaba's Qwen team put out open weights for its Qwen3.8 model family this week, releasing them under the Apache 2.0 license on Hugging Face and...