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。

Building the LLM OS by Andrej Karpathy with gpt-4o The speed + quality is 🔥🤯 code:

Ashpreet Bedi

19,960 subscribers

251,567 次观看 • 2 年前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

11 条评论

Ayush Chaurasia 的头像

Ayush Chaurasia2 年前

@karpathy Ashpreet’s routine: Wake up Drop a banger (Optionally) drop another banger Sleep 😴

Ashpreet Bedi 的头像

Ashpreet Bedi2 年前

@karpathy

Daniel Merja 的头像

Daniel Merja2 年前

@karpathy Add more RAM with MemGPT

Ali Madad 的头像

Ali Madad2 年前

@karpathy Feeling any qualitative difference from Llama-3 on Groq? Other than $$$.

Ashpreet Bedi 的头像

Ashpreet Bedi2 年前

@karpathy yes seems to be better than llama3, but a little below gpt-4-turbo

Sean 的头像

Sean2 年前

@karpathy @4thWaveStevie

Beach Please 的头像

Beach Please2 年前

@karpathy @ashpreetbedi do you post this on youtube? If not, please do?

Ashpreet Bedi 的头像

Ashpreet Bedi2 年前

@karpathy @B3achPleas3 youtube:

Yaya Soumah 的头像

Yaya Soumah2 年前

@karpathy You should make it into a dmg download, it could be easier to start using it.

Kiran Adimatyam 的头像

Kiran Adimatyam2 年前

@karpathy damn...how fast are you? I just finished your LLM OS video yesterday. Have the plan to try it today :) Thank you and Mr. Andrej for this thought. 4o cements that idea much more strongly.

Ashpreet Bedi 的头像

Ashpreet Bedi2 年前

@karpathy agree 4o adds the speed it was missing :) give it try, still a concept so would love to hear what you think

相关视频

Andrej Karpathy calls large language models the new computing paradigm: CPU -> LLM bytes -> tokens RAM -> context window this is the large language model OS (LMOS)

Andrej Karpathy calls large language models the new computing paradigm: CPU -> LLM bytes -> tokens RAM -> context window this is the large language model OS (LMOS)

ℏεsam

343,327 次观看 • 1 年前

The code for σ-GPT is out! Feel free to experiment with it here: It extends the François Fleuret picoclvr (non-nlp tasks) and Andrej Karpathy nanoGPT (text) codebases, with an additional KV-Cache from IRIS by Vincent Micheli and Eloi Alonso. (links below)

The code for σ-GPT is out! Feel free to experiment with it here: It extends the François Fleuret picoclvr (non-nlp tasks) and Andrej Karpathy nanoGPT (text) codebases, with an additional KV-Cache from IRIS by Vincent Micheli and Eloi Alonso. (links below)

Arnaud Pannatier

12,600 次观看 • 2 年前

Use GPT-4o for FREE with Bolt Crazy update just dropped. Don't want to pay for LLM API keys just to use Bolt? Now the Bolt 1-click launcher lets you use GPT-4o and GPT-4o-mini completely for FREE, powered by Github and LiteLLM (YC W23) - watch me get started in 1 minute:

Use GPT-4o for FREE with Bolt Crazy update just dropped. Don't want to pay for LLM API keys just to use Bolt? Now the Bolt 1-click launcher lets you use GPT-4o and GPT-4o-mini completely for FREE, powered by Github and LiteLLM (YC W23) - watch me get started in 1 minute:

cocktail peanut

128,085 次观看 • 1 年前

Interested in seeing how AI at Meta LLama 3.1 70B powered by Groq compares to OpenAI GPT-4o and GPT-4o Mini? We were too, so we decided to have them face off in the StreetFighter LLM Colosseum by Stan Girard and the team at Phospho.

Interested in seeing how AI at Meta LLama 3.1 70B powered by Groq compares to OpenAI GPT-4o and GPT-4o Mini? We were too, so we decided to have them face off in the StreetFighter LLM Colosseum by Stan Girard and the team at Phospho.

Groq Inc

179,931 次观看 • 2 年前

Introducing Predicted Outputs—dramatically decrease latency for gpt-4o and gpt-4o-mini by providing a reference string. Speed up: - Updating a blog post in a doc - Iterating on prior responses - Rewriting code in an existing file, like Exponent here:

Introducing Predicted Outputs—dramatically decrease latency for gpt-4o and gpt-4o-mini by providing a reference string. Speed up: - Updating a blog post in a doc - Iterating on prior responses - Rewriting code in an existing file, like Exponent here:

OpenAI Developers

580,764 次观看 • 1 年前

BREAKING: ChatGPT GPT-4o was just announce by OpenAI. It improves on vision, audio and text. The ease of use is incredibly enhanced. It makes interaction with the GPT much more natural, especially with voice. GPT-4o reasons across voice, text and vision. GPT-4 wil be available to everyone.

BREAKING: ChatGPT GPT-4o was just announce by OpenAI. It improves on vision, audio and text. The ease of use is incredibly enhanced. It makes interaction with the GPT much more natural, especially with voice. GPT-4o reasons across voice, text and vision. GPT-4 wil be available to everyone.

Ed Krassenstein

21,605 次观看 • 2 年前

The ChatGPT Mac app is the ultimate screenshot-to-code tool. Screenshot anything, paste it in the ChatGPT shortcut, and just tell GPT-4o to code it for you. Here's me taking a snapshot of Snake Game and getting fully working code in 90 seconds. Video is on 3x speed.

The ChatGPT Mac app is the ultimate screenshot-to-code tool. Screenshot anything, paste it in the ChatGPT shortcut, and just tell GPT-4o to code it for you. Here's me taking a snapshot of Snake Game and getting fully working code in 90 seconds. Video is on 3x speed.

Rowan Cheung

860,731 次观看 • 2 年前

RouteLLM - Automatically routes your query to the best LLM Combines the chops of o1, Sonnet 3.5, GPT-4o and Gemini! Super Intelligence in the making 😄

RouteLLM - Automatically routes your query to the best LLM Combines the chops of o1, Sonnet 3.5, GPT-4o and Gemini! Super Intelligence in the making 😄

Bindu Reddy

72,973 次观看 • 1 年前

Andrej Karpathy (Andrej Karpathy), OpenAI co-founder, ex-Tesla AI, "vibe coding" creator. In just 4 mins, he explains why Claude Skills, MCP servers, and AI agents are past the hype and are now the new baseline for building. Worth every second ↓

Andrej Karpathy (Andrej Karpathy), OpenAI co-founder, ex-Tesla AI, "vibe coding" creator. In just 4 mins, he explains why Claude Skills, MCP servers, and AI agents are past the hype and are now the new baseline for building. Worth every second ↓

Charly Wargnier

416,345 次观看 • 3 个月前

Andrej Karpathy Professor Tolkien on what The Lord of The Rings is all about:

Andrej Karpathy Professor Tolkien on what The Lord of The Rings is all about:

𝐑𝐮𝐠𝐠𝐚

13,821 次观看 • 11 个月前

🚨 MiniCPM-o 2.6, an 8B multimodal LLM, outperforms GPT-4o, Gemini 1.5 Pro, and Sonnet in visual (70.2 avg on OpenCompass, 1.8M pixel - 1344 x 1344), speech (bilingual realtime, beats GPT-4o-realtime in ASR/STT), with 75% fewer vision tokens, supporting 30+ languages 🔥

🚨 MiniCPM-o 2.6, an 8B multimodal LLM, outperforms GPT-4o, Gemini 1.5 Pro, and Sonnet in visual (70.2 avg on OpenCompass, 1.8M pixel - 1344 x 1344), speech (bilingual realtime, beats GPT-4o-realtime in ASR/STT), with 75% fewer vision tokens, supporting 30+ languages 🔥

Vaibhav (VB) Srivastav

42,009 次观看 • 1 年前

CUDA MODE hackathon today! Here's Andrej Karpathy on the 🏖️ origin story of llm.c, and what it hints at for the fast, simple, llm-compiled future of custom software.

CUDA MODE hackathon today! Here's Andrej Karpathy on the 🏖️ origin story of llm.c, and what it hints at for the fast, simple, llm-compiled future of custom software.

swyx

97,440 次观看 • 1 年前

That’s it. That’s the tweet. The Realtime API now supports WebRTC—you can add Realtime capabilities with just a handful of lines of code. We’ve also cut prices by 60%, added GPT-4o mini (10x cheaper than previous prices), improved voice quality, and made inputs more reliable.

That’s it. That’s the tweet. The Realtime API now supports WebRTC—you can add Realtime capabilities with just a handful of lines of code. We’ve also cut prices by 60%, added GPT-4o mini (10x cheaper than previous prices), improved voice quality, and made inputs more reliable.

OpenAI Developers

523,494 次观看 • 1 年前

LLM running on Windows 98 PC 26 year old hardware with Intel Pentium II CPU and 128MB RAM. Uses llama98.c, our custom pure C inference engine based on Andrej Karpathy llama2.c Code and DIY guide 👇

LLM running on Windows 98 PC 26 year old hardware with Intel Pentium II CPU and 128MB RAM. Uses llama98.c, our custom pure C inference engine based on Andrej Karpathy llama2.c Code and DIY guide 👇

EXO Labs

483,032 次观看 • 1 年前

The future that is happening. Automation powered by GPT-4o generates Figma designs based on PRD. Figma

The future that is happening. Automation powered by GPT-4o generates Figma designs based on PRD. Figma

yancymin

1,563,294 次观看 • 2 年前

ROUTE LLM - SUPER INTELLIGENCE IN THE MAKING RouteLLM combines the superpowers of o1-preview, Sonnet 3.5, Gemini Flash, o1-mini, 4o, and 4o-mini! Based on your query, it optimizes for latency, cost, and quality of the response. The cool thing about RouteLLM is it combines. - coding ability of Sonnet 3.5 - reasoning ability of o1-preview - hard-coding ability of o1-mini - 1M context of Gemini - speed of 4o-mini Similar to ensemble models that are very effective in classical ML, RouteLLM and similar models will be the most effective in the LLM ecosystem

ROUTE LLM - SUPER INTELLIGENCE IN THE MAKING RouteLLM combines the superpowers of o1-preview, Sonnet 3.5, Gemini Flash, o1-mini, 4o, and 4o-mini! Based on your query, it optimizes for latency, cost, and quality of the response. The cool thing about RouteLLM is it combines. - coding ability of Sonnet 3.5 - reasoning ability of o1-preview - hard-coding ability of o1-mini - 1M context of Gemini - speed of 4o-mini Similar to ensemble models that are very effective in classical ML, RouteLLM and similar models will be the most effective in the LLM ecosystem

Bindu Reddy

66,374 次观看 • 1 年前

Andrej @Karpathy explains on the Dwarkesh Patel podcast why he's not impressed by demos anymore Full episode:

Andrej @Karpathy explains on the Dwarkesh Patel podcast why he's not impressed by demos anymore Full episode:

Whole Mars Catalog

37,615 次观看 • 4 个月前