正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Introducing 𝗦𝘂𝗽𝗲𝗿 𝗝𝗦𝗢𝗡 𝗠𝗼𝗱𝗲, a framework for low latency structured output generation from LLMs. Generate JSON up to 𝟮𝟬𝘅 𝗳𝗮𝘀𝘁𝗲𝗿 from OpenAI and open source models. ❌ No need to threaten the model, tip the AI, etc ❌ Built with Alex Derhacobian 🔧 🧵👇

Varun Shenoy

6,930 subscribers

166,119 次观看 • 2 年前 •via X (Twitter)

科学技术新闻政治教育

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Introducing Inversion, our family of structured LLMs. Our first generation models excel in structured tasks, offering unmatched speed, latency, reliability, and efficiency, with the most comprehensive typed JSON output support available anywhere.

Introducing Inversion, our family of structured LLMs. Our first generation models excel in structured tasks, offering unmatched speed, latency, reliability, and efficiency, with the most comprehensive typed JSON output support available anywhere.

John

212,951 次观看 • 2 年前

Introducing the AI Registry Discover all models in the AI SDK AI Gateway: 🔍 Filter by: - Name - Input/output modalities - Context length - Output size - Providers 📄 View model details ⚖️ Compare model vs. model Open-source, link below!

Introducing the AI Registry Discover all models in the AI SDK AI Gateway: 🔍 Filter by: - Name - Input/output modalities - Context length - Output size - Providers 📄 View model details ⚖️ Compare model vs. model Open-source, link below!

Francisco Moretti

10,449 次观看 • 8 个月前

Introducing DeepThought-8B: Transparent reasoning model built on LLaMA-3.1 with test-time compute scaling. - JSON-structured thought chains & controllable inference paths. - ~16GB VRAM, competitive w/ 70B models. - Open model weights, and inference scripts.

Introducing DeepThought-8B: Transparent reasoning model built on LLaMA-3.1 with test-time compute scaling. - JSON-structured thought chains & controllable inference paths. - ~16GB VRAM, competitive w/ 70B models. - Open model weights, and inference scripts.

Ruliad

219,315 次观看 • 1 年前

LongWriter Unleashing 10,000+ Word Generation from Long Context LLMs discuss: Current long context large language models (LLMs) can process inputs up to 100,000 tokens, yet struggle to generate outputs exceeding even a modest length of 2,000 words. Through controlled experiments, we find that the model's effective generation length is inherently bounded by the sample it has seen during supervised fine-tuning (SFT). In other words, their output limitation is due to the scarcity of long-output examples in existing SFT datasets. To address this, we introduce AgentWrite, an agent-based pipeline that decomposes ultra-long generation tasks into subtasks, enabling off-the-shelf LLMs to generate coherent outputs exceeding 20,000 words. Leveraging AgentWrite, we construct LongWriter-6k, a dataset containing 6,000 SFT data with output lengths ranging from 2k to 32k words. By incorporating this dataset into model training, we successfully scale the output length of existing models to over 10,000 words while maintaining output quality. We also develop LongBench-Write, a comprehensive benchmark for evaluating ultra-long generation capabilities. Our 9B parameter model, further improved through DPO, achieves state-of-the-art performance on this benchmark, surpassing even much larger proprietary models. In general, our work demonstrates that existing long context LLM already possesses the potential for a larger output window--all you need is data with extended output during model alignment to unlock this capability.

LongWriter Unleashing 10,000+ Word Generation from Long Context LLMs discuss: Current long context large language models (LLMs) can process inputs up to 100,000 tokens, yet struggle to generate outputs exceeding even a modest length of 2,000 words. Through controlled experiments, we find that the model's effective generation length is inherently bounded by the sample it has seen during supervised fine-tuning (SFT). In other words, their output limitation is due to the scarcity of long-output examples in existing SFT datasets. To address this, we introduce AgentWrite, an agent-based pipeline that decomposes ultra-long generation tasks into subtasks, enabling off-the-shelf LLMs to generate coherent outputs exceeding 20,000 words. Leveraging AgentWrite, we construct LongWriter-6k, a dataset containing 6,000 SFT data with output lengths ranging from 2k to 32k words. By incorporating this dataset into model training, we successfully scale the output length of existing models to over 10,000 words while maintaining output quality. We also develop LongBench-Write, a comprehensive benchmark for evaluating ultra-long generation capabilities. Our 9B parameter model, further improved through DPO, achieves state-of-the-art performance on this benchmark, surpassing even much larger proprietary models. In general, our work demonstrates that existing long context LLM already possesses the potential for a larger output window--all you need is data with extended output during model alignment to unlock this capability.

AK

50,969 次观看 • 1 年前

Generate your own optimized cursor rule directly from your package.json, now live on Cursor Directory! Built using: ◇ Next.js - Framework ◇ Vercel - Hosting ◇ AI SDK - AI Toolkit ◇ xAI - LLM ◇ shadcn - UI Components Link and open source ⬇️🧵

Generate your own optimized cursor rule directly from your package.json, now live on Cursor Directory! Built using: ◇ Next.js - Framework ◇ Vercel - Hosting ◇ AI SDK - AI Toolkit ◇ xAI - LLM ◇ shadcn - UI Components Link and open source ⬇️🧵

Pontus Abrahamsson — oss/acc

115,522 次观看 • 1 年前

Want to generate images from the world's best models like Nano Banana or GPT Image side by side? Now you can with Image Lab 🚀 Run a single prompt across up to 25 models, with up to 20 images from each, and see results in seconds. You've seen our leaderboards. Now generate and evaluate the models yourself. Link in thread 🧵👇

Want to generate images from the world's best models like Nano Banana or GPT Image side by side? Now you can with Image Lab 🚀 Run a single prompt across up to 25 models, with up to 20 images from each, and see results in seconds. You've seen our leaderboards. Now generate and evaluate the models yourself. Link in thread 🧵👇

Artificial Analysis

12,250 次观看 • 4 个月前

Introducing $SAHARA — the native utility token of Sahara AI. Designed for real-world function, $SAHARA powers a new economic model for AI. One that is collaborative, open, and built to reward real contributors. Utility, distribution, and claim details below 🧵👇

Introducing $SAHARA — the native utility token of Sahara AI. Designed for real-world function, $SAHARA powers a new economic model for AI. One that is collaborative, open, and built to reward real contributors. Utility, distribution, and claim details below 🧵👇

Sahara AI 🔆

845,584 次观看 • 1 年前

Google just open-sourced LangExtract Python library! It uses LLMs to extract entities, attributes, and relations—with exact source grounding—from unstructured documents. Flexible LLM support (Gemini, OpenAI, Ollama) 100% open-source.

Google just open-sourced LangExtract Python library! It uses LLMs to extract entities, attributes, and relations—with exact source grounding—from unstructured documents. Flexible LLM support (Gemini, OpenAI, Ollama) 100% open-source.

Akshay 🚀

120,405 次观看 • 10 个月前

$1/ Introducing RWAi RWAi is the first platform where anyone can purchase fractional ownership and earn passive income in USDC from tokenized AI GPU rigs that run top open-source LLMs like DeepSeek and Llama at state of the art AI data centers🧵...$

1/ Introducing RWAi RWAi is the first platform where anyone can purchase fractional ownership and earn passive income in USDC from tokenized AI GPU rigs that run top open-source LLMs like DeepSeek and Llama at state of the art AI data centers🧵...

RWAi

16,108 次观看 • 1 年前

Introducing WELLAIOS: A revolutionary Open source AI Agent development framework and launchpad on Solana WELLAIOS empowers you to create, tokenize, and trade advanced AI agents on the Solana blockchain. Built by our team of AI professionals, our platform combines a sophisticated open-source framework with multi-modal AI technology, bringing character creation and storytelling to life without having to write a single line of code. Create intelligent digital characters that evolve, interact, and generate content through our library of AI agents. Each character becomes a unique asset equipped with a large set of AI tools on the Solana blockchain. For developers and creators: Join our thriving ecosystem where innovation meets rewards. Build plugins, contribute to our open-source framework, and earn as your creations shape the future. 👀 Sneak Peek: Watch our 3D AI agent capabilities showcase for content creation. Coming soon to Solana. Follow us on X for exclusive updates and be part of the revolution.

Introducing WELLAIOS: A revolutionary Open source AI Agent development framework and launchpad on Solana WELLAIOS empowers you to create, tokenize, and trade advanced AI agents on the Solana blockchain. Built by our team of AI professionals, our platform combines a sophisticated open-source framework with multi-modal AI technology, bringing character creation and storytelling to life without having to write a single line of code. Create intelligent digital characters that evolve, interact, and generate content through our library of AI agents. Each character becomes a unique asset equipped with a large set of AI tools on the Solana blockchain. For developers and creators: Join our thriving ecosystem where innovation meets rewards. Build plugins, contribute to our open-source framework, and earn as your creations shape the future. 👀 Sneak Peek: Watch our 3D AI agent capabilities showcase for content creation. Coming soon to Solana. Follow us on X for exclusive updates and be part of the revolution.

WELL3

10,503 次观看 • 1 年前

Ubisoft La Forge open-sourced its PBR material model CHORD, together with ComfyUI custom nodes for end-to-end AI material generation. This is a real step toward generative AI in AAA game pipelines and demonstrates how ComfyUI has grown from an open-source community tool to a production platform. Huge thanks to the Ubisoft La Forge team for giving back real production-grade tools to the open source community.

Ubisoft La Forge open-sourced its PBR material model CHORD, together with ComfyUI custom nodes for end-to-end AI material generation. This is a real step toward generative AI in AAA game pipelines and demonstrates how ComfyUI has grown from an open-source community tool to a production platform. Huge thanks to the Ubisoft La Forge team for giving back real production-grade tools to the open source community.

ComfyUI

96,065 次观看 • 6 个月前

We're excited to introduce Ocelot, an open-source and Brave-trained AI model built specifically to summarize web content. Ocelot is now available in Leo, our browser's AI assistant, and as an open-source model for the developer community.

We're excited to introduce Ocelot, an open-source and Brave-trained AI model built specifically to summarize web content. Ocelot is now available in Leo, our browser's AI assistant, and as an open-source model for the developer community.

Brave

100,685 次观看 • 2 个月前

HE FIXED A BROKEN $5 MP3 PLAYER WITH ONE PROMPT. SAME MODEL WITH OPENAI CODEX BUILT A CUSTOM OS his process: → bought cheap MP3 player on AliExpress → threw it away due to brutal Bluetooth stuttering → hit by a sudden shower thought → cracked open the case to check the chip → one year: garbage in a drawer the AI version: → showed the microchip to OpenAI Codex → Codex dumped firmware and analyzed binaries → built a custom OS fixing all Bluetooth bugs → remapped clunky navigation from left-right to up-down → result: instant, flawless performance the hardware is the same. the output is 10× bigger.

HE FIXED A BROKEN $5 MP3 PLAYER WITH ONE PROMPT. SAME MODEL WITH OPENAI CODEX BUILT A CUSTOM OS his process: → bought cheap MP3 player on AliExpress → threw it away due to brutal Bluetooth stuttering → hit by a sudden shower thought → cracked open the case to check the chip → one year: garbage in a drawer the AI version: → showed the microchip to OpenAI Codex → Codex dumped firmware and analyzed binaries → built a custom OS fixing all Bluetooth bugs → remapped clunky navigation from left-right to up-down → result: instant, flawless performance the hardware is the same. the output is 10× bigger.

leopardracer

15,479 次观看 • 29 天前

Each booth at Apsara 2025 is buzzing with the latest AI innovations! From the next-generation large language models of the Qwen3 family and the Wan 2.5 models to the new agent development framework feature in Model Studio, the updates are rolling in! Stay tuned as we zoom into the booths and bring the breakthroughs to you!🔥

Each booth at Apsara 2025 is buzzing with the latest AI innovations! From the next-generation large language models of the Qwen3 family and the Wan 2.5 models to the new agent development framework feature in Model Studio, the updates are rolling in! Stay tuned as we zoom into the booths and bring the breakthroughs to you!🔥

Alibaba Group

161,597 次观看 • 9 个月前

We built a better alternative to json-render & A2UI Up to 3x faster rendering & 67% fewer tokens. Far fewer broken outputs. We've been shipping Generative UI interfaces to 10,000+ developers for the past year. We know where JSON-based rendering breaks in production. So we built a new format - one that works with how models actually generate, not against it. And today we're open sourcing it: openui [dot] com

We built a better alternative to json-render & A2UI Up to 3x faster rendering & 67% fewer tokens. Far fewer broken outputs. We've been shipping Generative UI interfaces to 10,000+ developers for the past year. We know where JSON-based rendering breaks in production. So we built a new format - one that works with how models actually generate, not against it. And today we're open sourcing it: openui [dot] com

Rabi Shanker Guha

130,507 次观看 • 3 个月前

Introducing Arcana. The only framework that lets you build and run a complete onchain app in one place. Built for the way people build now. Prompt it, deploy it, run it onchain, all from one framework with no infrastructure to assemble.

Introducing Arcana. The only framework that lets you build and run a complete onchain app in one place. Built for the way people build now. Prompt it, deploy it, run it onchain, all from one framework with no infrastructure to assemble.

Arcana

72,984 次观看 • 1 个月前

The v0.14 release of BERTopic is here 🥳 Fine-tune your topic keywords and labels with models from OpenAI, Hugging Face, @CohereAI, spaCy, and . Use models for part-of-speech tagging, text generation, zero-shot classification, and more! An overview thread👇🧵

The v0.14 release of BERTopic is here 🥳 Fine-tune your topic keywords and labels with models from OpenAI, Hugging Face, @CohereAI, spaCy, and . Use models for part-of-speech tagging, text generation, zero-shot classification, and more! An overview thread👇🧵

Maarten Grootendorst

124,476 次观看 • 3 年前

here's how to do this in 2 minutes: - upload monk image to gemini - ask it to generate a 1:1 json prompt - generate the voice with minimax audio - generate the image with nano banana pro - lipsync with infinitetalk (open-source) congrats you just saved $10k for a trip to china to find a monk.

here's how to do this in 2 minutes: - upload monk image to gemini - ask it to generate a 1:1 json prompt - generate the voice with minimax audio - generate the image with nano banana pro - lipsync with infinitetalk (open-source) congrats you just saved $10k for a trip to china to find a monk.

0x ROAS

242,758 次观看 • 5 个月前

Introducing AI CLI Generate images, video, and text from your terminal. Pipe them together. Works with any agent. ai image "sunset" | ai video "animate" → Hundreds of models → Multi-model comparison → Inline previews, no native deps → AI SDK + AI Gateway

Introducing AI CLI Generate images, video, and text from your terminal. Pipe them together. Works with any agent. ai image "sunset" | ai video "animate" → Hundreds of models → Multi-model comparison → Inline previews, no native deps → AI SDK + AI Gateway

Chris Tate

61,385 次观看 • 2 个月前

Up to 3x faster LLM generation with no extra resources/requirements - ngram speculation has landed in 🤗 transformers! 🏎️💨 All you need to do is to add `prompt_lookup_num_tokens=10` to your `generate` call and you'll get faster LLMs🔥 🧵How does it work?

Up to 3x faster LLM generation with no extra resources/requirements - ngram speculation has landed in 🤗 transformers! 🏎️💨 All you need to do is to add `prompt_lookup_num_tokens=10` to your `generate` call and you'll get faster LLMs🔥 🧵How does it work?

João Gante

107,741 次观看 • 2 年前