Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Open models can come from anywhere. Securing them shouldn’t depend on where you get them. Today we're launching Project Beacon to make safety a default for open intelligence. We're partnering with Goodfire to bring inference and in-line monitoring together for the first time, so safety signals are caught proactively.... show more

Baseten

21,435 subscribers

22,103 views • 3 days ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

4 Comments

Mark Attar3 days ago

@GoodfireAI Yo @oneill_c you lost the beard

Amelia | AI & Tech3 days ago

@GoodfireAI In-line monitoring making safety the default for open models is a solid step enough

Sonup Bhuyaan3 days ago

@GoodfireAI Safety by default at the inference layer makes sense. Most teams pulling open models never add monitoring on their own.

Sam3 days ago

@GoodfireAI "Safety as a default" is doing a lot of work in that sentence. What does the default actually block, and can a customer turn it off?

Related Videos

Until now, adding web search to open-source models meant hand-wiring orchestration, managing separate keys, and paying latency taxes on every round trip. Today, we're solving that. We're excited to introduce Baseten Hosted Tools and Baseten Grounded Inference to bring real-time web search server-side to open models running on Baseten through a single configuration: - 15% lower latency compared to client-side execution - No extra vendor key required - Zero orchestration We're launching this preview version with four leading web search partners, Exa, Keenable AI, Parallel Web Systems, and You.com. Get frontier-level web search parity for your open-weight models. More here:

Until now, adding web search to open-source models meant hand-wiring orchestration, managing separate keys, and paying latency taxes on every round trip. Today, we're solving that. We're excited to introduce Baseten Hosted Tools and Baseten Grounded Inference to bring real-time web search server-side to open models running on Baseten through a single configuration: - 15% lower latency compared to client-side execution - No extra vendor key required - Zero orchestration We're launching this preview version with four leading web search partners, Exa, Keenable AI, Parallel Web Systems, and You.com. Get frontier-level web search parity for your open-weight models. More here:

Baseten

29,259 views • 26 days ago

We're rounding off our #JingleJam2024 charity intros with The Trevor Project—we're so glad to be partnering with them for the very first time. Support The Trevor Project and aid their work to end suicide among LGBTQ+ young people through suicide prevention and crisis intervention.

We're rounding off our #JingleJam2024 charity intros with The Trevor Project—we're so glad to be partnering with them for the very first time. Support The Trevor Project and aid their work to end suicide among LGBTQ+ young people through suicide prevention and crisis intervention.

Jingle Jam

11,322 views • 1 year ago

🚀 Our "technical" marketer might not be looped in, but today is our biggest launch day yet. We're introducing two new products to serve the inference lifecycle: Model APIs and Training. Model APIs are frontier models running on the Baseten Inference Stack, purpose-built for production. Baseten Training (Beta) provides infra and tooling without limitations for AI models destined for production. Huge shoutout to the many partners and customers we've worked with as we built these two new products—more details below.

🚀 Our "technical" marketer might not be looped in, but today is our biggest launch day yet. We're introducing two new products to serve the inference lifecycle: Model APIs and Training. Model APIs are frontier models running on the Baseten Inference Stack, purpose-built for production. Baseten Training (Beta) provides infra and tooling without limitations for AI models destined for production. Huge shoutout to the many partners and customers we've worked with as we built these two new products—more details below.

Baseten

35,358 views • 1 year ago

BOST ENERGIES X BUZSTOPBOYS: SAFETY, SERVICE & IMPACT! 🚧🤝 We're proud to partner with BOST Energies as part of their Health & Safety Week celebrations! Together, we're taking the message of health, safety, and environmental responsibility beyond the workplace and into our communities. This partnership reflects our shared commitment to creating cleaner, safer spaces while inspiring collective action for positive change. A big thank you to BOST Energies for joining hands with us to make a meaningful impact. Here's to building a culture where safety, service, and community always come first.

BOST ENERGIES X BUZSTOPBOYS: SAFETY, SERVICE & IMPACT! 🚧🤝 We're proud to partner with BOST Energies as part of their Health & Safety Week celebrations! Together, we're taking the message of health, safety, and environmental responsibility beyond the workplace and into our communities. This partnership reflects our shared commitment to creating cleaner, safer spaces while inspiring collective action for positive change. A big thank you to BOST Energies for joining hands with us to make a meaningful impact. Here's to building a culture where safety, service, and community always come first.

Buzstopboys

18,415 views • 3 months ago

Detroit Lions HC Dan Campbell - Meeting Structure "First of all I know this, you can give them deadhead if you're not careful, so hit them in spurts. You give them 20 minutes, hit them on what you need to hit them on & get them out for a couple minutes, bring them back & keep going." "It needs to be interactive, ultimately we're teachers. We're teachers in football, so how do you communicate & articulate what you're trying to get across to those players. Don't make it more complex than it needs to be, keep it simple where they can play fast, know what to do & make it interactive, make it fun.

Detroit Lions HC Dan Campbell - Meeting Structure "First of all I know this, you can give them deadhead if you're not careful, so hit them in spurts. You give them 20 minutes, hit them on what you need to hit them on & get them out for a couple minutes, bring them back & keep going." "It needs to be interactive, ultimately we're teachers. We're teachers in football, so how do you communicate & articulate what you're trying to get across to those players. Don't make it more complex than it needs to be, keep it simple where they can play fast, know what to do & make it interactive, make it fun.

James Light

299,275 views • 8 months ago

today we're launching opencode zen zen provides the best LLMs for coding ALWAYS in the best possible configuration no more "did claude get dumber?" no more being routed to a crappy provider and we're giving it to you at cost and you can use it anywhere that enough for you?

today we're launching opencode zen zen provides the best LLMs for coding ALWAYS in the best possible configuration no more "did claude get dumber?" no more being routed to a crappy provider and we're giving it to you at cost and you can use it anywhere that enough for you?

dax

199,474 views • 1 year ago

We're proud to be the fastest inference provider on Artificial Analysis, OpenRouter, and Hugging Face for GLM-5.3-Flash, at 122+ TPS. All served from the US only, starting on day 0, with ZDR by default. Stay tuned for updates as our engineers continue to optimize GLM-5.3-Flash for even higher throughput and lower latency.

We're proud to be the fastest inference provider on Artificial Analysis, OpenRouter, and Hugging Face for GLM-5.3-Flash, at 122+ TPS. All served from the US only, starting on day 0, with ZDR by default. Stay tuned for updates as our engineers continue to optimize GLM-5.3-Flash for even higher throughput and lower latency.

Baseten

23,114 views • 1 month ago

We're in a very narrow window of time right now where the very best technology is available to you. Frontier labs and open models are delivering state-of-the-art capability to anyone willing to build. But security, safety, and rapid self-improvement will inevitably change what that access looks like. The edge is available today, and windows like this don’t stay open for very long.

We're in a very narrow window of time right now where the very best technology is available to you. Frontier labs and open models are delivering state-of-the-art capability to anyone willing to build. But security, safety, and rapid self-improvement will inevitably change what that access looks like. The edge is available today, and windows like this don’t stay open for very long.

Dave Blundin

27,081 views • 7 months ago

Today we're launching a new way to build apps in AI Studio! 🚀 - Build directly in AI Studio and leverage Gemini without a key, for free! - Browse new sample apps that you can remix and make your own - Share with friends Check out the open source apps below, more to come!

Today we're launching a new way to build apps in AI Studio! 🚀 - Build directly in AI Studio and leverage Gemini without a key, for free! - Browse new sample apps that you can remix and make your own - Share with friends Check out the open source apps below, more to come!

Ammaar Reshi

60,758 views • 1 year ago

🇺🇸🇸🇴TRUMP: "CONVICTED NATURALIZED IMMIGRANTS, WE’RE GOING TO GET THEM THE HELL OUT OF HERE" "We're also going to revoke the citizenship of any naturalized immigrant from Somalia or anywhere else who is convicted of defrauding our citizens. We're going to get them the hell out of here first. And if you come to America to rob Americans, we're throwing you in jail and we're sending you back to the place from where you came. We're throwing you right in jail." Source: Donald J. Trump, Eric Daugherty

🇺🇸🇸🇴TRUMP: "CONVICTED NATURALIZED IMMIGRANTS, WE’RE GOING TO GET THEM THE HELL OUT OF HERE" "We're also going to revoke the citizenship of any naturalized immigrant from Somalia or anywhere else who is convicted of defrauding our citizens. We're going to get them the hell out of here first. And if you come to America to rob Americans, we're throwing you in jail and we're sending you back to the place from where you came. We're throwing you right in jail." Source: Donald J. Trump, Eric Daugherty

Mario Nawfal

107,443 views • 9 months ago

Today, we're launching Belvedir We're making private AI models cheap and easy so the world can have trillions of them. Private AI models are cheaper, better, faster, and protect your data from frontier labs, but they're still to expensive and difficult to make for startups and consumers. Belvedir makes private AI models autonomously with less than 5 minutes of setup, so that private AI can finally be accessible to everyone

Today, we're launching Belvedir We're making private AI models cheap and easy so the world can have trillions of them. Private AI models are cheaper, better, faster, and protect your data from frontier labs, but they're still to expensive and difficult to make for startups and consumers. Belvedir makes private AI models autonomously with less than 5 minutes of setup, so that private AI can finally be accessible to everyone

Zachary Yu

90,793 views • 2 months ago

AI has two sides: the applications that need intelligence, and the computers that provide it. OptimAI Router brings them closer together. For developers: access a growing range of open AI models through one connection. For node operators: turn your computer into a potential source of AI inference, contributing computing resources to real workloads across the network. One side brings the demand. The other brings the capacity. OptimAI connects them. We're building toward an AI network where developers can access diverse models, more computing resources can participate, and intelligence isn't tied to a single provider. Build with open models. Contribute computing power. Help grow the network. One network. From AI requests to distributed compute. OptimAI Router. Coming soon. ⚡

AI has two sides: the applications that need intelligence, and the computers that provide it. OptimAI Router brings them closer together. For developers: access a growing range of open AI models through one connection. For node operators: turn your computer into a potential source of AI inference, contributing computing resources to real workloads across the network. One side brings the demand. The other brings the capacity. OptimAI connects them. We're building toward an AI network where developers can access diverse models, more computing resources can participate, and intelligence isn't tied to a single provider. Build with open models. Contribute computing power. Help grow the network. One network. From AI requests to distributed compute. OptimAI Router. Coming soon. ⚡

OptimAI Network

14,104 views • 11 days ago

This scene hit hard and more polls show it's unfortunately accurate sadly. "They let us all come out, so now we're standing in the open ready for them to shoot us down" - such a devastating line.

This scene hit hard and more polls show it's unfortunately accurate sadly. "They let us all come out, so now we're standing in the open ready for them to shoot us down" - such a devastating line.

BUILD MORE HOUSING!

219,498 views • 4 months ago

Rivian CEO RJ Scaringe in new interview on self-driving: "Tesla's got a huge fleet. So our approach is we have to catch up with them and compete. And so we're going to have more cameras with better dynamic range. We're going to supplement that with LiDAR that provides better safety for edge case conditions and allows us to train the vision models faster. And the incremental cost to do that is relatively small. And in fact any of the incremental costs that would have been there on its own is offset by the fact that we brought inference in house and reduced the cost of our inference platform so dramatically from what we have in your car. Your car uses an Nvidia inference platform. I say all this because in like the infinite long-term you know, you could make the case that once the models are very very robust you could have less cameras or you may be able to get away with less radar. It's not clear yet if that's the case for covering all these corner cases."

Rivian CEO RJ Scaringe in new interview on self-driving: "Tesla's got a huge fleet. So our approach is we have to catch up with them and compete. And so we're going to have more cameras with better dynamic range. We're going to supplement that with LiDAR that provides better safety for edge case conditions and allows us to train the vision models faster. And the incremental cost to do that is relatively small. And in fact any of the incremental costs that would have been there on its own is offset by the fact that we brought inference in house and reduced the cost of our inference platform so dramatically from what we have in your car. Your car uses an Nvidia inference platform. I say all this because in like the infinite long-term you know, you could make the case that once the models are very very robust you could have less cameras or you may be able to get away with less radar. It's not clear yet if that's the case for covering all these corner cases."

Sawyer Merritt

451,115 views • 7 months ago

AVs on Uber's platform see 30% higher utilization... In 2020, Uber sold its self-driving unit. Since then, they've become the biggest aggregator of AV supply, partnering with nearly every AV provider (eg Waymo). Dara explains their strategy: "We're allowing these companies to really focus on building the driver. We are building services around them. We're securing depots, charging, and working with fleet partners. We're securing financing. We just announced a billion-dollar financing line with Santander for EV fleets and AV fleets. We're working on autonomous insurance as well. We're building the entire ecosystem so someone who builds a driver can get those cars in the road, and then, we have all the supporting infrastructure. When they hit market, we've got instant demand for them. AVs that are on our network are 30% or more busy than 1P AVs who aren't using our network. A Waymo, of course they want to build their own brand and their own channel, but at the same time we are able to drive more utilization for them when their cars are on our network. We think there's going to be an amalgamation of business models. There already is in travel, there already is in food — same thing will happen in transportation."

Patrick OShaughnessy

45,488 views • 4 months ago

Today we're announcing Base Labs, a dedicated research organization focused on advancing open-source AI. We believe in a healthy, open frontier model ecosystem. To enable this, we are working on: - Blue-sky research on continual learning, the science of RL, and how models learn across their full lifecycle, with every experiment and recipe shared openly. - The BaseHub Data Foundry: the highest-quality open RL environments, training data, and real-world benchmarks, built for anyone to train and benchmark on. - Post-post training: taking open-source models and making them better, safer, and more aligned through continual post-training, built on our research, and deployed with our frontier safety stack so organizations can use open-source models with confidence. - Making models cheaper and more performant through our model performance research. This is a mission-driven research effort, not a commercial product. We believe the health of the open-source AI ecosystem matters and that the best way to advance it is to do science in the open. We’re hiring engineers, researchers, and research fellows to advance this mission.

Today we're announcing Base Labs, a dedicated research organization focused on advancing open-source AI. We believe in a healthy, open frontier model ecosystem. To enable this, we are working on: - Blue-sky research on continual learning, the science of RL, and how models learn across their full lifecycle, with every experiment and recipe shared openly. - The BaseHub Data Foundry: the highest-quality open RL environments, training data, and real-world benchmarks, built for anyone to train and benchmark on. - Post-post training: taking open-source models and making them better, safer, and more aligned through continual post-training, built on our research, and deployed with our frontier safety stack so organizations can use open-source models with confidence. - Making models cheaper and more performant through our model performance research. This is a mission-driven research effort, not a commercial product. We believe the health of the open-source AI ecosystem matters and that the best way to advance it is to do science in the open. We’re hiring engineers, researchers, and research fellows to advance this mission.

Base Labs

244,598 views • 1 month ago

We're open! For the first time you can get a schooner of Saison de Maison direct from the brewery. So much work has gone into today but this makes it so worth it. We're so thankful to everyone that has helped us get to this point. Now, it's time to enjoy a couple of beers! 🍻

We're open! For the first time you can get a schooner of Saison de Maison direct from the brewery. So much work has gone into today but this makes it so worth it. We're so thankful to everyone that has helped us get to this point. Now, it's time to enjoy a couple of beers! 🍻

Balance Brewing & Blending

11,166 views • 3 years ago

President of the International Longshoremen’s Association, Harold J. Daggett, 100% means business: "People never gave a s*** about us until now. When they finally realized that the chain is being broke now ... You know how many people depend on our jobs. Half the world! It's time for them and time for Washington to put so much pressure on them to take care of us because we took care of them. And we're here 135 years and brought them where they are today and they don't want to share. We're done." Things are about to get wild.

President of the International Longshoremen’s Association, Harold J. Daggett, 100% means business: "People never gave a s*** about us until now. When they finally realized that the chain is being broke now ... You know how many people depend on our jobs. Half the world! It's time for them and time for Washington to put so much pressure on them to take care of us because we took care of them. And we're here 135 years and brought them where they are today and they don't want to share. We're done." Things are about to get wild.

Charlie Kirk

5,833,292 views • 2 years ago

Journalist, "Some tourists will need to provide five years of social media to get a tourist visa. Are you concerned it will lead to a decline in tourism?" President Trump, "No, we're doing so well. We just want people to come over and know its safe" "We want safety, we want security, we want to make sure we're not letting the wrong people come to our country" "We're getting rid of a lot of bad people from jail, from mental institutions, drug dealers. We're getting a lot of bad people, murderers" "11,888 murders, almost half of them committed more than one murder. We're getting rid of them"

Journalist, "Some tourists will need to provide five years of social media to get a tourist visa. Are you concerned it will lead to a decline in tourism?" President Trump, "No, we're doing so well. We just want people to come over and know its safe" "We want safety, we want security, we want to make sure we're not letting the wrong people come to our country" "We're getting rid of a lot of bad people from jail, from mental institutions, drug dealers. We're getting a lot of bad people, murderers" "11,888 murders, almost half of them committed more than one murder. We're getting rid of them"

Farrukh

438,176 views • 10 months ago