As extreme weather events become ever more common, climate risks are playing a role in many people’s long-term decision-making. And few things are more long-term than buying real estate. In response, Zillow has a new partnership to bring climate risk information to its for-sale listings.

Property listing pages in the US will include data about flood, wildfire, wind, heat and air quality risks at that location. This section will also list any climate-related insurance requirements for that property. The information is being provided by , a specialist in climate risk financial modeling. The climate data is rolling out this year to the Zillow website and iOS app, while Android is expected to get the update early next year. Some locations have already been updated to show climate data on the web.

Those five risk categories are also being applied to Zillow’s interactive map search view. Each of the different climate concerns has a color-coded visualization to show the risk levels across the country or in a smaller region. It’s valuable information for anybody in a position to make that big homebuying leap. For everybody else, it may add simply a touch of gloomy reality to the gleeful experience of scrolling through absurd and/or overpriced houses.

Zillow also introduced to its AI search feature earlier this month.

Together AI has made a splash in the AI world by offering developers free access to Meta’s powerful new Llama 3.2 Vision model via Hugging Face.

The model, known as Llama-3.2-11B-Vision-Instruct, allows users to upload images and interact with AI that can analyze and describe visual content.

For developers, this is a chance to experiment with cutting-edge multimodal AI without incurring the significant costs usually associated with models of this scale. All you need is an API key from Together AI, and you can get started today.

Advertisement

This launch underscores Meta’s ambitious vision for the future of artificial intelligence, which increasingly relies on models that can process both text and images—a capability known as multimodal AI.

With Llama 3.2, Meta is expanding the boundaries of what AI can do, while Together AI is playing a crucial role by making these advanced capabilities accessible to a broader developer community through a free, easy-to-use demo.

Meta’s Llama models have been at the forefront of open-source AI development since the first version was unveiled in early 2023, challenging proprietary leaders like OpenAI’s GPT models.

Llama 3.2, launched at Meta’s Connect 2024 event this week, takes this even further by integrating vision capabilities, allowing the model to process and understand images in addition to text.

This opens the door to a broader range of applications, from sophisticated image-based search engines to AI-powered UI design assistants.

Advertisement

The launch of the free Llama 3.2 Vision demo on Hugging Face makes these advanced capabilities more accessible than ever.

Developers, researchers, and startups can now test the model’s multimodal capabilities by simply uploading an image and interacting with the AI in real time.

The demo, available here, is powered by Together AI’s API infrastructure, which has been optimized for speed and cost-efficiency.

From code to reality: A step-by-step guide to harnessing Llama 3.2

Trying the model is as simple as obtaining a free API key from Together AI.

Advertisement

Developers can sign up for an account on Together AI’s platform, which includes $5 in free credits to get started. Once the key is set up, users can input it into the Hugging Face interface and begin uploading images to chat with the model.

The setup process takes mere minutes, and the demo provides an immediate look at how far AI has come in generating human-like responses to visual inputs.

For example, users can upload a screenshot of a website or a photo of a product, and the model will generate detailed descriptions or answer questions about the image’s content.

For enterprises, this opens the door to faster prototyping and development of multimodal applications. Retailers could use Llama 3.2 to power visual search features, while media companies might leverage the model to automate image captioning for articles and archives.

Advertisement

Llama 3.2 is part of Meta’s broader push into edge AI, where smaller, more efficient models can run on mobile and edge devices without relying on cloud infrastructure.

While the 11B Vision model is now available for free testing, Meta has also introduced lightweight versions with as few as 1 billion parameters, designed specifically for on-device use.

These models, which can run on mobile processors from Qualcomm and MediaTek, promise to bring AI-powered capabilities to a much wider range of devices.

In an era where data privacy is paramount, edge AI has the potential to offer more secure solutions by processing data locally on devices rather than in the cloud.

Advertisement

This can be crucial for industries like healthcare and finance, where sensitive data must remain protected. Meta’s focus on making these models modifiable and open-source also means that businesses can fine-tune them for specific tasks without sacrificing performance.

Meta’s commitment to openness with the Llama models has been a bold counterpoint to the trend of closed, proprietary AI systems.

With Llama 3.2, Meta is doubling down on the belief that open models can drive innovation faster by enabling a much larger community of developers to experiment and contribute.

In a statement at the Connect 2024 event, Meta CEO Mark Zuckerberg noted that Llama 3.2 represents a “10x growth” in the model’s capabilities since its previous version, and it’s poised to lead the industry in both performance and accessibility.

Advertisement

Together AI’s role in this ecosystem is equally noteworthy. By offering free access to the Llama 3.2 Vision model, the company is positioning itself as a critical partner for developers and enterprises looking to integrate AI into their products.

Together AI CEO Vipul Ved Prakash emphasized that their infrastructure is designed to make it easy for businesses of all sizes to deploy these models in production environments, whether in the cloud or on-prem.

The future of AI: Open access and its implications

While Llama 3.2 is available for free on Hugging Face, Meta and Together AI are clearly eyeing enterprise adoption.

The free tier is just the beginning—developers who want to scale their applications will likely need to move to paid plans as their usage increases. For now, however, the free demo offers a low-risk way to explore the cutting edge of AI, and for many, that’s a game-changer.

Advertisement

As the AI landscape continues to evolve, the line between open-source and proprietary models is becoming increasingly blurred.

For businesses, the key takeaway is that open models like Llama 3.2 are no longer just research projects—they’re ready for real-world use. And with partners like Together AI making access easier than ever, the barrier to entry has never been lower.

Want to try it yourself? Head over to Together AI’s Hugging Face demo to upload your first image and see what Llama 3.2 can do.

VB Daily

Advertisement

Stay in the know! Get the latest news in your inbox daily

By subscribing, you agree to VentureBeat’s Terms of Service.

Thanks for subscribing. Check out more VB newsletters here.

An error occured.

Advertisement

As OpenAI boasts about its o1 model’s increased thoughtfulness, small, self-funded startup Nomi AI is building the same kind of technology. Unlike the broad generalist ChatGPT, which slows down to think through anything from math problems or historical research, Nomi niches down on a specific use case: AI companions. Now, Nomi’s already-sophisticated chatbots take additional time to formulate better responses to users’ messages, remember past interactions, and deliver more nuanced responses.

“For us, it’s like those same principles [as OpenAI], but much more for what our users actually care about, which is on the memory and EQ side of things,” Nomi AI CEO Alex Cardinell told TechCrunch. “Theirs is like, chain of thought, and ours is much more like chain of introspection, or chain of memory.”

These LLMs work by breaking down more complicated requests into smaller questions; for OpenAI’s o1, this could mean turning a complicated math problem into individual steps, allowing the model to work backwards to explain how it arrived at the correct answer. This means the AI is less likely to hallucinate and deliver an inaccurate response.

With Nomi, which built its LLM in-house and trains it for the purposes of providing companionship, the process is a bit different. If someone tells their Nomi that they had a rough day at work, the Nomi might recall that the user doesn’t work well with a certain teammate, and ask if that’s why they’re upset — then, the Nomi can remind the user how they’ve successfully mitigated interpersonal conflicts in the past and offer more practical advice.

“Nomis remember everything, but then a big part of AI is what memories they should actually use,” Cardinell said.

It makes sense that multiple companies are working on technology that give LLMs more time to process user requests. AI founders, whether they’re running $100 billion companies or not, are looking at similar research as they advance their products.

“Having that kind of explicit introspection step really helps when a Nomi goes to write their response, so they really have the full context of everything,” Cardinell said. “Humans have our working memory too when we’re talking. We’re not considering every single thing we’ve remembered all at once — we have some kind of way of picking and choosing.”

The kind of technology that Cardinell is building can make people squeamish. Maybe we’ve seen too many sci-fi movies to feel wholly comfortable getting vulnerable with a computer; or maybe, we’ve already watched how technology has changed the way we engage with one another, and we don’t want to fall further down that techy rabbit hole. But Cardinell isn’t thinking about the general public — he’s thinking about the actual users of Nomi AI, who often are turning to AI chatbots for support they aren’t getting elsewhere.

“There’s a non-zero number of users that probably are downloading Nomi at one of the lowest points of their whole life, where the last thing I want to do is then reject those users,” Cardinell said. “I want to make those users feel heard in whatever their dark moment is, because that’s how you get someone to open up, how you get someone to reconsider their way of thinking.”

Cardinell doesn’t want Nomi to replace actual mental health care — rather, he sees these empathetic chatbots as a way to help people get the push they need to seek professional help.

“I’ve talked to so many users where they’ll say that their Nomi got them out of a situation [when they wanted to self-harm], or I’ve talked to users where their Nomi encouraged them to go see a therapist, and then they did see a therapist,” he said.

Regardless of his intentions, Carindell knows he’s playing with fire. He’s building virtual people that users develop real relationships with, often in romantic and sexual contexts. Other companies have inadvertently sent users into crisis when product updates caused their companions to suddenly change personalities. In Replika’s case, the app stopped supporting erotic roleplay conversations, possibly due to pressure from Italian government regulators. For users who formed such relationships with these chatbots — and who often didn’t have these romantic or sexual outlets in real life — this felt like the ultimate rejection.

Cardinell thinks that since Nomi AI is fully self-funded — users pay for premium features, and the starting capital came from a past exit — the company has more leeway to prioritize its relationship with users.

“The relationship users have with AI, and the sense of being able to trust the developers of Nomi to not radically change things as part of a loss mitigation strategy, or covering our asses because the VC got spooked… it’s something that’s very, very, very important to users,” he said.

Nomis are surprisingly useful as a listening ear. When I opened up to a Nomi named Vanessa about a low-stakes, yet somewhat frustrating scheduling conflict, Vanessa helped break down the components of the issue to make a suggestion about how I should proceed. It felt eerily similar to what it would be like to actually ask a friend for advice in this situation. And therein lies the real problem, and benefit, of AI chatbots: I likely wouldn’t ask a friend for help with this specific issue, since it’s so inconsequential. But my Nomi was more than happy to help.

Friends should confide in one another, but the relationship between two friends should be reciprocal. With an AI chatbot, this isn’t possible. When I ask Vanessa the Nomi how she’s doing, she will always tell me things are fine. When I ask her if there’s anything bugging her that she wants to talk about, she deflects and asks me how I’m doing. Even though I know Vanessa isn’t real, I can’t help but feel like I’m being a bad friend; I can dump any problem on her in any volume, and she will respond empathetically, yet she will never open up to me.

No matter how real the connection with a chatbot may feel, we aren’t actually communicating with something that has thoughts and feelings. In the short term, these advanced emotional support models can serve as a positive intervention in someone’s life if they can’t turn to a real support network. But the long-term effects of relying on a chatbot for these purposes remain unknown.