In the AI world, there’s a buzz in the air about a new AI language model released Tuesday by Meta: Llama 3.1 405B. The reason? It’s potentially the first time anyone can download a GPT-4-class large language model (LLM) for free and run it on their own hardware. You’ll still need some beefy hardware: Meta says it can run on a "single server node," which isn’t desktop PC-grade equipment. But it’s a provocative shot across the bow of "closed" AI model vendors such as OpenAI and Anthropic.

"Llama 3.1 405B is the first openly available model that rivals the top AI models when it comes to state-of-the-art capabilities in general knowledge, steerability, math, tool use, and multilingual translation," says Meta. Company CEO Mark Zuckerberg calls 405B "the first frontier-level open source AI model."

In the AI industry, "frontier model" is a term for an AI system designed to push the boundaries of current capabilities. In this case, Meta is positioning 405B among the likes of the industry’s top AI models, such as OpenAI’s GPT-4o, Claude’s 3.5 Sonnet, and Google Gemini 1.5 Pro.

A chart published by Meta suggests that 405B gets very close to matching the performance of GPT-4 Turbo, GPT-4o, and Claude 3.5 Sonnet in benchmarks like MMLU (undergraduate level knowledge), GSM8K (grade school math), and HumanEval (coding).

But as we’ve noted many times since March, these benchmarks aren’t necessarily scientifically sound or translate to the subjective experience of interacting with AI language models. In fact, this traditional slate of AI benchmarks is so generally useless to laypeople that even Meta’s PR department now just posts a few images of charts and doesn’t even try to explain them in any detail.

We’ve instead found that measuring the subjective experience of using a conversational AI model (through what might be called "vibemarking") on A/B leaderboards like Chatbot Arena is a better way to judge new LLMs. In the absence of Chatbot Arena data, Meta has provided the results of its own human evaluations of 405B’s outputs that seem to show Meta’s new model holding its own against GPT-4 Turbo and Claude 3.5 Sonnet.

Whatever the benchmarks, early word on the street (after the model leaked on 4chan yesterday) seems to match the claim that 405B is roughly equivalent to GPT-4. It took a lot of expensive computer training time to get there—and money, of which the social media giant has plenty to burn. Meta trained the 405B model on over 15 trillion tokens of training data scraped from the web (then parsed, filtered, and annotated by Llama 2), using more than 16,000 H100 GPUs.

So what’s with the 405B name? In this case, "405B" means 405 billion parameters, and parameters are numerical values that store trained information in a neural network. More parameters translate to a larger neural network powering the AI model, which generally (but not always) means more capability, such as better ability to make contextual connections between concepts. But larger-parameter models have a tradeoff in needing more computing power (AKA "compute") to run.

We’ve been expecting the release of a 400+ billion parameter model of the Llama 3 family since Meta gave word that it was training one in April, and today’s announcement isn’t just about the biggest member of the Llama 3 family: There’s an entire new iteration of improved Llama models with the designation "Llama 3.1." That includes upgraded versions of its smaller 8B and 70B models, which now feature multilingual support and an extended context length of 128,000 tokens (the "context length" is roughly the working memory capacity of the model, and "tokens" are chunks of data used by LLMs to process information).

Meta says that 405B is useful for long-form text summarization, multilingual conversational agents, and coding assistants and for creating synthetic data used to train future AI language models. Notably, that last use-case—allowing developers to use outputs from Llama models to improve other AI models—is now officially supported by Meta’s Llama 3.1 license for the first time.

Abusing the term “open source”

Llama 3.1 405B is an open-weights model, which means anyone can download the trained neural network files and run them or fine-tune them. That directly challenges a business model where companies like OpenAI keep the weights to themselves and instead monetize the model through subscription wrappers like ChatGPT or charge for access by the token through an API.

Fighting the "closed" AI model is a big deal to Mark Zuckerberg, who simultaneously released a 2,300-word manifesto today on why the company believes in open releases of AI models, titled, "Open Source AI Is the Path Forward." More on the terminology in a minute. But briefly, he writes about the need for customizable AI models that offer user control and encourage better data security, higher cost-efficiency, and better future-proofing, as opposed to vendor-locked solutions.

All that sounds reasonable, but undermining your competitors using a model subsidized by a social media war chest is also an efficient way to play spoiler in a market where you might not always win with the most cutting-edge tech. That benefits Meta, Zuckerberg says, because he doesn’t want to get locked into a system where companies like his have to pay a toll to access AI capabilities, drawing comparisons to "taxes" Apple levies on developers through its App Store.

So about that "open source" term. As we first wrote in an update to our Llama 2 launch article a year ago, "open source" has a very particular meaning that has traditionally been defined by the Open Source Initiative. The AI industry has not yet settled on terminology for AI model releases that ship either code or weights with restrictions (such as Llama 3.1) or that ship without providing training data. We’ve been calling these releases "open weights" instead.

Unfortunately for terminology sticklers, Zuckerberg has now baked the erroneous "open source" label into the title of his potentially historic aforementioned essay on open AI releases, so fighting for the correct term in AI may be a losing battle. Still, his usage annoys people like independent AI researcher Simon Willison, who likes Zuckerberg’s essay otherwise.

"I see Zuck’s prominent misuse of ‘open source’ as a small-scale act of cultural vandalism," Willison told Ars Technica. "Open source should have an agreed meaning. Abusing the term weakens that meaning which makes the term less generally useful, because if someone says ‘it’s open source,’ that no longer tells me anything useful. I have to then dig in and figure out what they’re actually talking about."

The Llama 3.1 models are available for download through Meta’s own website and on Hugging Face. They both require providing contact information and agreeing to a license and an acceptable use policy, which means that Meta can technically legally pull the rug out from under your use of Llama 3.1 or its outputs at any time.

Fans of director Denis Villeneuve’s epic two-part film adaptation of Frank Herbert’s Dune have no doubt been curious about the upcoming HBO Max series, Dune: Prophecy. It’s a prequel series inspired by the novel Sisterhood of Dune, written by Brian Herbert and Kevin J. Anderson, exploring the origins of the Bene Gesserit. The studio just dropped a tantalizing teaser rife with political intrigue, ominous warnings, and a bit of hand-to-hand combat.

The series was first announced in 2019, with Villeneuve serving as an executive producer and Alison Schapker (Alias, Fringe, Altered Carbon) serving as showrunner. The first season will consist of six episodes, and it’s unclear how closely the series will adhere to the source material. Per the official premise:

Set 10,000 years before the ascension of Paul Atreides, Dune: Prophecy follows two Harkonnen sisters as they combat forces that threaten the future of humankind, and establish the fabled sect that will become known as the Bene Gesserit.

Emily Watson co-stars as Valya Harkonnen, leader of the Sisterhood, with Olivia Williams playing her sister, Tula Harkonnen. Mark Strong plays Emperor Javicco Corrino, described as "a man from a great line of war-time Emperors, who is called upon to govern the Imperium and manage a fragile peace," while Jodhi May plays Empress Natalya and Sarah-Sofie Boussnina plays Princess Ynez.

The cast also includes Shalom Brune-Franklin as Mikaela, a Fremen woman who serves the royal family; Travis Fimmel as Desmond Hart, described as "a charismatic soldier with an enigmatic past"; Chris Mason as swordsman Keiran Atreides; Josh Heuston as Constantine Corrino, the illegitimate son of Javicco; Edward Davis as rising politician Harrow Harkonnen; Tabu as Sister Francesca, the Emperor’s former lover; Jihae as Reverend Mother Kasha, the Emperor’s Truthsayer; Faoileann Cunningham as Sister Jen, Chloe Lea as Lila, Jade Anouka as Sister Theodosia, and Aoife Hinds as Sister Emeline, all acolytes at the Sisterhood School.

Power = control

A short teaser was shown in May during the Warner Bros. Discovery Upfront presentation in New York City. It was heavy on the exposition, with a voiceover describing the founding of a sisterhood assigned to the Great Houses "to help them sift truth from lies." The result was a "network of influence throughout the Imperium… but power comes with a price."  They want to place a Sister on the throne and arrange a marriage to make it possible. Not all the Sisters were on board with the plan, however, with one warning that the Sisterhood was playing god "and we will be judged for it."

This latest teaser opens with an admonition to acolytes of the Sisterhood: "You wish to serve the Great Houses and shape the flow of power; you must first exert power over yourself." The emperor seems to be easily wooed by the "sorceresses," much to his empress’s chagrin, but the more influence the Sisterhood wields, the more enemies it gains. Desmond Hart also has his suspicions about the Sisterhood, probably with good reason. "Our hands are poised on the levers of power but yet our grasp on it is still fragile," Valya tells her sister Tula, assuring her that "I am trying to protect the Imperium"—and "sacrifices must be made."

Dune: Prophecy premieres this November on Max.

Listing image by HBO Max