正在加载视频...

视频加载失败

返回首页

Introducing Synthesis! We built a new tool that merges different model responses into one:

charlieholtz's profile picture

Charlie Holtz

36,467 subscribers

78,901 次观看 • 1 年前 •via X (Twitter)

教育艺术科学技术
LIVE

Anya Rossi• Live Now

Private livecam show

11 条评论

Charlie Holtz 的头像
Charlie Holtz1 年前

You can try it here:

Charlie Holtz 的头像
Charlie Holtz1 年前

A big part of the inspiration for synthesis was this post by @karpathy: and the blog post in the video is here!

The Information 的头像
The Information1 年前

The best tools for sensibly introducing kids to AI.

Erik Dunteman 的头像
Erik Dunteman1 年前

Hell yeah, fellow launcher!

Reza Sayar 的头像
Reza Sayar1 年前

this is super cool! 🔥🔥

Tair Asim 的头像
Tair Asim1 年前

fire ship. choosing a model for synthesizing in interesting. how many levels deep it can go with combining diff models 🤔

Vamsi Batchu 的头像
Vamsi Batchu1 年前

Dude you are cooking!!!

Alex Gracian 的头像
Alex Gracian1 年前

Just tried it. Noticed that when you hit the synthesis button it give you the combined output, but you lose all the individual results for each model. Anyway, of still keeping the original outputs above the synthesis result?

Web3 的头像
Web31 年前

This is a game changer. Nice work, would love to test this out for my business.

Harry Fletcher-Jones (👨‍🎨,👨‍🎨) 的头像
Harry Fletcher-Jones (👨‍🎨,👨‍🎨)1 年前

This is cool Which model performs the synthesis?

Charlie Holtz 的头像
Charlie Holtz1 年前

Thanks! Right now Claude 3-5 sonnet, but likely making it configurable soon

相关视频

Introducing WebSockets in the Responses API. Built for low-latency, long-running agents with heavy tool calls.
0:22

Introducing WebSockets in the Responses API. Built for low-latency, long-running agents with heavy tool calls.

OpenAI Developers

1,347,280 次观看 • 6 个月前

Introducing Claude 3.7 Sonnet: our most intelligent model to date. It's a hybrid reasoning model, producing near-instant responses or extended, step-by-step thinking. One model, two ways to think. We’re also releasing an agentic coding tool: Claude Code.
0:39

Introducing Claude 3.7 Sonnet: our most intelligent model to date. It's a hybrid reasoning model, producing near-instant responses or extended, step-by-step thinking. One model, two ways to think. We’re also releasing an agentic coding tool: Claude Code.

Anthropic

3,627,062 次观看 • 1 年前

Most active projects carry a few memories that quietly contradict each other. We built something for that! Introducing Dream, the background layer that keeps mem0 memory clean, current, and true. → Merge: facts that repeat get folded into one clear memory → Supersede: the old fact stays as history, the new fact becomes the truth → Synthesis: not one memory — a pattern built from many, traced back to each one Excited to share that we're live today on Pro and Enterprise!
1:00

Most active projects carry a few memories that quietly contradict each other. We built something for that! Introducing Dream, the background layer that keeps mem0 memory clean, current, and true. → Merge: facts that repeat get folded into one clear memory → Supersede: the old fact stays as history, the new fact becomes the truth → Synthesis: not one memory — a pattern built from many, traced back to each one Excited to share that we're live today on Pro and Enterprise!

mem0

10,558 次观看 • 1 个月前

Introducing Empathic Voice Interface 2 (EVI 2), our new voice-to-voice foundation model. EVI 2 merges language and voice into a single model trained specifically for emotional intelligence. You can try it and start building today.
8:06

Introducing Empathic Voice Interface 2 (EVI 2), our new voice-to-voice foundation model. EVI 2 merges language and voice into a single model trained specifically for emotional intelligence. You can try it and start building today.

Hume AI

165,645 次观看 • 2 年前

🎉We founded Meta Pi 🔥a business platform that merges a new subscription economy model with the innovative digital economy of #Web3.
1:42

🎉We founded Meta Pi 🔥a business platform that merges a new subscription economy model with the innovative digital economy of #Web3.

Meta Pi

16,585 次观看 • 2 年前

🔊New NVIDIA paper: Audio-SDS🔊 We repurpose Score Distillation Sampling (SDS) for audio, turning any pretrained audio diffusion model into a tool for diverse tasks, including source separation, impact synthesis & more. 🎧 Demos, audio examples, paper:
1:54

🔊New NVIDIA paper: Audio-SDS🔊 We repurpose Score Distillation Sampling (SDS) for audio, turning any pretrained audio diffusion model into a tool for diverse tasks, including source separation, impact synthesis & more. 🎧 Demos, audio examples, paper:

Jonathan Lorraine

39,443 次观看 • 1 年前

DigitalOcean’s model synthesis tool is now available in public preview. 🆕 A new server-side tool: run multiple models on one inference request and get a single synthesized response, no custom orchestration. Just add it to your existing inference request.
3:22

DigitalOcean’s model synthesis tool is now available in public preview. 🆕 A new server-side tool: run multiple models on one inference request and get a single synthesized response, no custom orchestration. Just add it to your existing inference request.

DigitalOcean

2,192,210 次观看 • 1 个月前

Introducing Critique, a new multi-model deep research system in M365 Copilot. You can use multiple models together to generate optimal responses and reports.
0:50

Introducing Critique, a new multi-model deep research system in M365 Copilot. You can use multiple models together to generate optimal responses and reports.

Satya Nadella

1,393,245 次观看 • 5 个月前

From NYC, we are building something different. A new kind of financial system, built for everyone. That’s money built different.
1:04

From NYC, we are building something different. A new kind of financial system, built for everyone. That’s money built different.

Circle

65,661 次观看 • 1 年前

Introducing Whisper Thunder aka Gen-4.5. Today, we are excited to share our new frontier model. Gen-4.5 was built by a team that fits onto two school buses and decided to take on the largest companies in the world. We are David and we’ve brought one hell of a slingshot.
2:07

Introducing Whisper Thunder aka Gen-4.5. Today, we are excited to share our new frontier model. Gen-4.5 was built by a team that fits onto two school buses and decided to take on the largest companies in the world. We are David and we’ve brought one hell of a slingshot.

Nicolas Neubert

702,637 次观看 • 9 个月前

Introducing LTX-2.5, the world model the world builds on. One of the biggest upgrades yet to the model already powering film, robotics, and real-time workflows. Higher pixel fidelity, multishot scenes that hold together across cuts, and a pretrained foundation built to be finetuned across domains. Plus: Diffusion Fidelity Rendering, a new rendering approach built for pixel quality that holds up frame by frame, even on a cinema screen. This isn't a tool you rent. It's a foundation you build on, open and yours.
2:03

Introducing LTX-2.5, the world model the world builds on. One of the biggest upgrades yet to the model already powering film, robotics, and real-time workflows. Higher pixel fidelity, multishot scenes that hold together across cuts, and a pretrained foundation built to be finetuned across domains. Plus: Diffusion Fidelity Rendering, a new rendering approach built for pixel quality that holds up frame by frame, even on a cinema screen. This isn't a tool you rent. It's a foundation you build on, open and yours.

LTX.io

13,727,565 次观看 • 1 个月前

Introducing 3e8blueprint. A tool for turning an Idea into hardware.
0:38

Introducing 3e8blueprint. A tool for turning an Idea into hardware.

David Feldt

548,963 次观看 • 8 个月前

Oda was so nasty for building up Aramaki just to turn him into a Shanks hype tool. Yonko are built different 😭
1:35

Oda was so nasty for building up Aramaki just to turn him into a Shanks hype tool. Yonko are built different 😭

𝓩𝓸𝓻𝓸 ゾロ

39,921 次观看 • 4 个月前

Novel view synthesis has long been a core challenge in 3D vision. But how much 3D inductive bias is truly needed? —Surprisingly, very little! Introducing "LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias"—a fully transformer-based approach that enables scalable, generalizable, and fully data-driven novel view synthesis, from sparse posed inputs. 🧵(1/6) Project Page:
0:36

Novel view synthesis has long been a core challenge in 3D vision. But how much 3D inductive bias is truly needed? —Surprisingly, very little! Introducing "LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias"—a fully transformer-based approach that enables scalable, generalizable, and fully data-driven novel view synthesis, from sparse posed inputs. 🧵(1/6) Project Page:

Haian Jin

114,937 次观看 • 1 年前

GPT-6 Astra built a full 3D model of the human cell in 30 minutes. Biology didn't see this coming. Not a diagram. Not an illustration. A complete, detailed model built from scratch in a single session. The same tool that handles code, design, and 3D geometry is now walking into science.
0:45

GPT-6 Astra built a full 3D model of the human cell in 30 minutes. Biology didn't see this coming. Not a diagram. Not an illustration. A complete, detailed model built from scratch in a single session. The same tool that handles code, design, and 3D geometry is now walking into science.

Zentrix⌚️

51,436 次观看 • 6 天前

Introducing StyleDrop, a model that allows a significantly higher level of stylized text-to-image synthesis by using a few style reference images that describe the style for text-to-image generation, bypassing the burden of text prompt engineering. More→
0:42

Introducing StyleDrop, a model that allows a significantly higher level of stylized text-to-image synthesis by using a few style reference images that describe the style for text-to-image generation, bypassing the burden of text prompt engineering. More→

Google AI

80,379 次观看 • 2 年前

introducing Context for Nano Banana. we created a brand new tool for Nano Banana Pro that makes it easy to add context to your generations. learn more 👇
0:50

introducing Context for Nano Banana. we created a brand new tool for Nano Banana Pro that makes it easy to add context to your generations. learn more 👇

Krea

122,123 次观看 • 9 个月前

Can a language model become SOTA in chemical synthesis? 🧪🔥 We put a 2.6B-parameter Liquid AI model into MMAI Gym. It now sets the state of the art in single-step retrosynthesis. 🚀 #insilicoSOTAFM
0:20

Can a language model become SOTA in chemical synthesis? 🧪🔥 We put a 2.6B-parameter Liquid AI model into MMAI Gym. It now sets the state of the art in single-step retrosynthesis. 🚀 #insilicoSOTAFM

Insilico Medicine

37,422 次观看 • 11 天前

we've been disappointed at how the input into most apps is still just mouse and keyboard. So we built a tool that lets you write directly into any app - check it out
0:36

we've been disappointed at how the input into most apps is still just mouse and keyboard. So we built a tool that lets you write directly into any app - check it out

WillScuderi

3,741,838 次观看 • 2 年前

we've been disappointed at how the input into most apps is still just mouse and keyboard. So we built a tool that lets you write directly into any app - check it out
0:23

we've been disappointed at how the input into most apps is still just mouse and keyboard. So we built a tool that lets you write directly into any app - check it out

WillScuderi

5,826,984 次观看 • 2 年前