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。

I just fine-tuned Google DeepMind's "gemma-3-270m" and it's ready for secure deployment in production! → Imported and fine-tuned using Hugging Face transformers on M4 → Packaged and versioned with KitOps → Pushed to Jozu Hub for production use and deployments All within just a few minutes🔥 "gemma-3-270m" is lightweight,... show more

Mr. Ånand

2,030 subscribers

53,839 次观看 • 1 年前 •via X (Twitter)

科学技术教育

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

Few seconds and <4GB to full fine tune Gemma 3 270M for classification tasks! Here on M3 Ultra! I'll keep experimenting and checking final result! base model, batch size 4, adaw (testing with adafactor too!), rank 128. Video at normal speed here! 🔥

Few seconds and <4GB to full fine tune Gemma 3 270M for classification tasks! Here on M3 Ultra! I'll keep experimenting and checking final result! base model, batch size 4, adaw (testing with adafactor too!), rank 128. Video at normal speed here! 🔥

Ivan Fioravanti ᯅ

35,282 次观看 • 1 年前

Google engineer explained how to fine-tune a tiny LLM from 46% to 90% accuracy on your phone in 21 minutes - better than $1500 on-device AI bootcamps. pick Gemma 270M -> generate synthetic task data -> fine-tune with LoRA -> quantize to int4 -> deploy to Pixel and hit 2000 tokens per second. That loop is how a 270M model beats a 70B one on your task, running fully offline in your pocket. Gemma 270M + synthetic data + LoRA + int4 quantization + on-device runtime - that's the stack. Watch and save it, then fine-tune your own tiny agent tonight.

Google engineer explained how to fine-tune a tiny LLM from 46% to 90% accuracy on your phone in 21 minutes - better than $1500 on-device AI bootcamps. pick Gemma 270M -> generate synthetic task data -> fine-tune with LoRA -> quantize to int4 -> deploy to Pixel and hit 2000 tokens per second. That loop is how a 270M model beats a 70B one on your task, running fully offline in your pocket. Gemma 270M + synthetic data + LoRA + int4 quantization + on-device runtime - that's the stack. Watch and save it, then fine-tune your own tiny agent tonight.

h100envy

706,789 次观看 • 1 个月前

🚨BREAKING: Every vibe coding startup just had a very bad week. Google just shipped production-grade full-stack coding for free. Google AI Studio just went full-stack, and it's designed to turn your prompts into production-ready apps, Here’s what actually dropped 👇

🚨BREAKING: Every vibe coding startup just had a very bad week. Google just shipped production-grade full-stack coding for free. Google AI Studio just went full-stack, and it's designed to turn your prompts into production-ready apps, Here’s what actually dropped 👇

Poonam Soni

88,861 次观看 • 5 个月前

🔥 We fine-tuned Gemma 3 using Unsloth AI on a custom instruction dataset scraped and built with . The model is now able to answer questions about OpenAI Agents SDK documentation. Check out the demo below👇

🔥 We fine-tuned Gemma 3 using Unsloth AI on a custom instruction dataset scraped and built with . The model is now able to answer questions about OpenAI Agents SDK documentation. Check out the demo below👇

Eric Ciarla (hiring)

28,794 次观看 • 1 年前

🚨 NOW: SecWar Pete Hegseth has just struck MAJOR AGREEMENTS to TRIPLE patriot (PAC-3 MSE) interceptor production capacity — and QUADRUPLE THAAD interceptor production capacity This is crucial for our troops! 🔥🔥🔥 7-year deals just signed with General Dynamics Ordnance and Tactical Systems, and Lockheed Martin This will also surge manufacturing for "essential sub-components, including motor cases, seeker housings and midsections, and shroud deployment systems" Pete Hegseth 🇺🇸

🚨 NOW: SecWar Pete Hegseth has just struck MAJOR AGREEMENTS to TRIPLE patriot (PAC-3 MSE) interceptor production capacity — and QUADRUPLE THAAD interceptor production capacity This is crucial for our troops! 🔥🔥🔥 7-year deals just signed with General Dynamics Ordnance and Tactical Systems, and Lockheed Martin This will also surge manufacturing for "essential sub-components, including motor cases, seeker housings and midsections, and shroud deployment systems" Pete Hegseth 🇺🇸

Eric Daugherty

77,668 次观看 • 3 天前

🚀 OpenLLM 0.6 is out providing fast inference speed! Check out our video comparing #OpenLLM and #Ollama handling concurrent requests on the Llama 3 8B model. Ollama is ideal for local LLM deployment, but it is not designed for high concurrency scenarios essential for deployments in the cloud, powering applications with a large number of users. OpenLLM enables you to run any open-source model as OpenAI-compatible endpoints with just a single command. It supports a broad array of models like Llama 3, Mistral, and Qwen 2, including those fine-tuned with proprietary datasets and quantized versions. Ready to speed up your AI deployments? See the OpenLLM repo #BentoML #OpenSource

🚀 OpenLLM 0.6 is out providing fast inference speed! Check out our video comparing #OpenLLM and #Ollama handling concurrent requests on the Llama 3 8B model. Ollama is ideal for local LLM deployment, but it is not designed for high concurrency scenarios essential for deployments in the cloud, powering applications with a large number of users. OpenLLM enables you to run any open-source model as OpenAI-compatible endpoints with just a single command. It supports a broad array of models like Llama 3, Mistral, and Qwen 2, including those fine-tuned with proprietary datasets and quantized versions. Ready to speed up your AI deployments? See the OpenLLM repo #BentoML #OpenSource

BentoML

18,798 次观看 • 2 年前

Everyone can build a chatbot. Not everyone can ship a secure, production-grade agent. In Episode 3 of AI Guide to the Galaxy, we break down how to enable your AI agents for the real-world deployments with Docker’s MCP Toolkit and Gateway. Watch here:

Everyone can build a chatbot. Not everyone can ship a secure, production-grade agent. In Episode 3 of AI Guide to the Galaxy, we break down how to enable your AI agents for the real-world deployments with Docker’s MCP Toolkit and Gateway. Watch here:

Docker

48,500 次观看 • 1 年前

Introducing react-native-wallet, built in collaboration with Expensify! 🚀 💳 A better way to add payment cards to Apple & Google Wallet 🔁 Truly cross-platform support for iOS & Android ⚡ Lightweight, dev-friendly, and easy to integrate 🔐 Secure, compliant, and production-ready in-app provisioning

Introducing react-native-wallet, built in collaboration with Expensify! 🚀 💳 A better way to add payment cards to Apple & Google Wallet 🔁 Truly cross-platform support for iOS & Android ⚡ Lightweight, dev-friendly, and easy to integrate 🔐 Secure, compliant, and production-ready in-app provisioning

Software Mansion

10,508 次观看 • 1 年前

Google released MedGemma on I/O'25 👏 > 4B and 27B instruction fine-tuned vision LMs and a 4B pre-trained vision LM for medicine > available with transformers from the get-go 🤗 they also released a cool demo for scan reading ⤵️

Google released MedGemma on I/O'25 👏 > 4B and 27B instruction fine-tuned vision LMs and a 4B pre-trained vision LM for medicine > available with transformers from the get-go 🤗 they also released a cool demo for scan reading ⤵️

merve

81,353 次观看 • 1 年前

Prophet Suddenly 3 is set to hit screens within the next two months! Filmed across Kenya, the United States, and Nigeria, this powerful project has been in production since last year, and we couldn’t be more thrilled. More than just a film, Prophet Suddenly 3 serves as God’s mouthpiece to our generation, carrying a message that is both timely and transformative. Stay tuned. #prophetsuddenly3 #miokukusheshe #prophetsuddenlyiscoming

Prophet Suddenly 3 is set to hit screens within the next two months! Filmed across Kenya, the United States, and Nigeria, this powerful project has been in production since last year, and we couldn’t be more thrilled. More than just a film, Prophet Suddenly 3 serves as God’s mouthpiece to our generation, carrying a message that is both timely and transformative. Stay tuned. #prophetsuddenly3 #miokukusheshe #prophetsuddenlyiscoming

The Winlos

43,624 次观看 • 1 年前

Is AI video at production quality yet? We asked Mason Warner, creative director at Slash: "Seedance 2.0 (ByteDance) is truly phenomenal." "I think it's at production quality. We've been using it for our last five major launches and advertisements." "Whatever production budgets I conceptualize in my head ($20,000-$100,000), I can do it for ~$1,000 to $2,000 of credits and 18 hours of just sitting on my computer."

Is AI video at production quality yet? We asked Mason Warner, creative director at Slash: "Seedance 2.0 (ByteDance) is truly phenomenal." "I think it's at production quality. We've been using it for our last five major launches and advertisements." "Whatever production budgets I conceptualize in my head ($20,000-$100,000), I can do it for ~$1,000 to $2,000 of credits and 18 hours of just sitting on my computer."

MTS

14,118 次观看 • 2 个月前

✈️ Designed to end reliance on Boeing and Airbus, Russia’s new MS-21 passenger jet just flew for 75 minutes with fully domestic systems onboard — no imports. But the real challenge? Mass production. You can’t replace Western planes with prototypes.

✈️ Designed to end reliance on Boeing and Airbus, Russia’s new MS-21 passenger jet just flew for 75 minutes with fully domestic systems onboard — no imports. But the real challenge? Mass production. You can’t replace Western planes with prototypes.

Brian McDonald

461,934 次观看 • 1 年前

Final validation is a group effort! These four R2 pre-production vehicles are just a few of the hundreds currently accumulating miles on roads and highways all over the country as we prepare for full-scale production in 2026.

Final validation is a group effort! These four R2 pre-production vehicles are just a few of the hundreds currently accumulating miles on roads and highways all over the country as we prepare for full-scale production in 2026.

Rivian

88,810 次观看 • 11 个月前

Whew! Big work today getting optimized llama.cpp MIPS R8000 kernel running on the SGI Power Challenge deskside from 1995 with Gemma 3 270M. Went from 0.16 to 0.5 tok/s with a combination of MIPSPro compiler and manual asm optimizations. Not bad for 30 year old hardware. 1/

Whew! Big work today getting optimized llama.cpp MIPS R8000 kernel running on the SGI Power Challenge deskside from 1995 with Gemma 3 270M. Went from 0.16 to 0.5 tok/s with a combination of MIPSPro compiler and manual asm optimizations. Not bad for 30 year old hardware. 1/

Nathan Odle

47,289 次观看 • 4 个月前

omg… Google’s AI Studio can create AI-powered apps ready for production in less than 5 minutes. I built this entire app using Gemini 3 without writing a single line of code. (And it’s free)

omg… Google’s AI Studio can create AI-powered apps ready for production in less than 5 minutes. I built this entire app using Gemini 3 without writing a single line of code. (And it’s free)

Poonam Soni

42,709 次观看 • 9 个月前

Retool just went from powerful to mind-blowing. The new `Assist` feature lets you build full, production-ready apps from a single prompt 🤯 Trigger agents, connect data, and deploy, all with enterprise-grade security. Let’s dive in 🧵↓

Retool just went from powerful to mind-blowing. The new `Assist` feature lets you build full, production-ready apps from a single prompt 🤯 Trigger agents, connect data, and deploy, all with enterprise-grade security. Let’s dive in 🧵↓

Charly Wargnier

59,389 次观看 • 11 个月前

how to reverse engineer any video and copy it's style, composition and scenes: i've said it before and i'll said it again... video production is the biggest opportunity in AI and Higgsfield just handed you the blueprint to make your own films these guys spent $500k in tokens to produce a full movie somehow they managed to nail character consistency, lip sync and motion for 95min and now the whole workflow and production methods behind it is free for anyone to use

how to reverse engineer any video and copy it's style, composition and scenes: i've said it before and i'll said it again... video production is the biggest opportunity in AI and Higgsfield just handed you the blueprint to make your own films these guys spent $500k in tokens to produce a full movie somehow they managed to nail character consistency, lip sync and motion for 95min and now the whole workflow and production methods behind it is free for anyone to use

Machina

20,509 次观看 • 1 个月前

You can now fine-tune Llama 3 without writing a single line of code! We are moving at breakneck speed. I recorded a video to show you how to fine-tune any open-source model in a few minutes. I'm using a GPT capable of taking a problem and turning it into a fine-tuned model that will solve it. You don't have to write any code. You only need to explain to a GPT what problem you want to solve and tell it you want to use Llama 3. For example, "fine-tune Llama 3" or "deploy zephyr." It feels magic. The system will recommend a dataset and fine-tune the model for you. I'm using Monster API, a platform that specializes in making fine-tuning and deploying open-source models easy and fast. Their stack is well-optimized to maximize fine-tuning efficiency using techniques like Q-Lora and vLLM. They are behind the GPT. Here is what you need to do: 1. Create an account at 2. Load the GPT with the link below This is as simple as it gets. When you are done, you can click a button to deploy the model and start using it. I have 10,000 free credits for anyone using the code "SANTIAGO" in the dashboard. You can use these credits to access, fine-tune, and deploy these open-source models. You can also keep up with their latest updates, and get free credits and special offers on their Discord server:

You can now fine-tune Llama 3 without writing a single line of code! We are moving at breakneck speed. I recorded a video to show you how to fine-tune any open-source model in a few minutes. I'm using a GPT capable of taking a problem and turning it into a fine-tuned model that will solve it. You don't have to write any code. You only need to explain to a GPT what problem you want to solve and tell it you want to use Llama 3. For example, "fine-tune Llama 3" or "deploy zephyr." It feels magic. The system will recommend a dataset and fine-tune the model for you. I'm using Monster API, a platform that specializes in making fine-tuning and deploying open-source models easy and fast. Their stack is well-optimized to maximize fine-tuning efficiency using techniques like Q-Lora and vLLM. They are behind the GPT. Here is what you need to do: 1. Create an account at 2. Load the GPT with the link below This is as simple as it gets. When you are done, you can click a button to deploy the model and start using it. I have 10,000 free credits for anyone using the code "SANTIAGO" in the dashboard. You can use these credits to access, fine-tune, and deploy these open-source models. You can also keep up with their latest updates, and get free credits and special offers on their Discord server:

Santiago

324,621 次观看 • 2 年前

Igbemo Ekiti Farm Cluster. Land cleared: 10ha. Located very close to a big water pool. The land is divided into: 1. Horticultural production through the use of screen houses and open field methods. 2. Fish farming: 4 big earthen ponds have been constructed. 3. On and off season maize farming with the use of surface and underground water sources. You can utilise our farm clusters initiative to undertake any agricultural production. We have a cluster in Ilasa Ekiti that is 100% dedicated to cocoa production by the promoters. Ekiti Food production systems!

Igbemo Ekiti Farm Cluster. Land cleared: 10ha. Located very close to a big water pool. The land is divided into: 1. Horticultural production through the use of screen houses and open field methods. 2. Fish farming: 4 big earthen ponds have been constructed. 3. On and off season maize farming with the use of surface and underground water sources. You can utilise our farm clusters initiative to undertake any agricultural production. We have a cluster in Ilasa Ekiti that is 100% dedicated to cocoa production by the promoters. Ekiti Food production systems!

Ebenezer Boluwade

40,015 次观看 • 1 年前