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.

This is wild! Llama 3.1 405B Instruct finally solves a famous math puzzle that was originally posted on /LocalLlama. To the best of my knowledge, every model (including Claude 3.5 Sonnet and GPT-4o) fails at this task. A longer video coming soon!

elvis

318,142 subscribers

52,808 views • 2 years ago •via X (Twitter)

Gaming Science & Technology Education

Anya Rossi• Live Now

Private livecam show

11 Comments

elvis2 years ago

Full video with an overview and test cases of the Llama 3.1 release:

RedDeer.Games5 years ago

Bring Your Switch To Life With The "Triple-A Clock" #AAAClock ⬅️ ➡️ #OLED Ready #NintendoSwitch

Juggalos For Context🌴🥥2 years ago

Was the answer in the training data?

elvis2 years ago

Could be! I need to try different variants as I mentioned in the video.

明2 years ago

@aidan_mclau

Lior⚡2 years ago

Huge!

Jasper2 years ago

We supported Llama 3.1 405B model on our distributed GPU network at @hyperbolic_labs. Come and use the API for FREE at

Shadow2 years ago

It is wild, man 🔥

Sivaram Raju2 years ago

That's incredible! 🎉

Skandarux.py2 years ago

Maybe it was included in training data ?

Andres Gomez Sarmiento2 years ago

Interesting test, but Claude Sonnet got it right for me:

Related Videos

Wow DeepSeek R1 version 1.5B runs perfectly locally on my phone 😳 So you can have a model that outperforms GPT-4o and Claude 3.5 Sonnet on math in your pocket. Mind-blowing

Wow DeepSeek R1 version 1.5B runs perfectly locally on my phone 😳 So you can have a model that outperforms GPT-4o and Claude 3.5 Sonnet on math in your pocket. Mind-blowing

Paul Couvert

665,544 views • 1 year ago

We’ve never seen a 7B model get the famous Sally test right, let alone a 3.8B model. 💪Even the Llama 3 8B Instruct model fails this test. Coming 🔜 to your 📱

We’ve never seen a 7B model get the famous Sally test right, let alone a 3.8B model. 💪Even the Llama 3 8B Instruct model fails this test. Coming 🔜 to your 📱

Private LLM

36,972 views • 2 years ago

We’re excited to introduce Previews, a new feature that lets you see and interact with web applications generated directly in chats on Poe. Previews works particularly well with LLMs that excel at coding, including Claude 3.5 Sonnet, GPT-4o, and Gemini 1.5 Pro. (1/5)

We’re excited to introduce Previews, a new feature that lets you see and interact with web applications generated directly in chats on Poe. Previews works particularly well with LLMs that excel at coding, including Claude 3.5 Sonnet, GPT-4o, and Gemini 1.5 Pro. (1/5)

Poe

116,311 views • 2 years ago

This is the fastest I've seen Llama 3.3 running anywhere! Llama 3.3 70B running at 652 t/s is lightning fast. And if you want Llama 3.1, here are the speeds I was able to get: • Llama 3.1 8B: 1006 t/s • Llama 3.1 70B: 709 t/s • Llama 3.1 405B: 206 t/s (You can access all of these models for free! See the link below.) This speed is incredible, but the interesting part is what's happening behind the scenes: These models aren't running on a GPU! In this video, I'm using the SambaNova cloud to access these models. They built a custom chip (SN40L) optimized for AI workflows. A single SN40L chip can hold hundreds of models (trillions of parameters) in memory! The speed alone is a huge deal, but the big advantage is for agentic workflows running multiple specialized models. A GPU can only host a single model and switch (unload and load) to a different model if necessary. An SN40L, on the other hand, can host every model at once, making it much faster. Here is the video where you can see how fast these chips are:

This is the fastest I've seen Llama 3.3 running anywhere! Llama 3.3 70B running at 652 t/s is lightning fast. And if you want Llama 3.1, here are the speeds I was able to get: • Llama 3.1 8B: 1006 t/s • Llama 3.1 70B: 709 t/s • Llama 3.1 405B: 206 t/s (You can access all of these models for free! See the link below.) This speed is incredible, but the interesting part is what's happening behind the scenes: These models aren't running on a GPU! In this video, I'm using the SambaNova cloud to access these models. They built a custom chip (SN40L) optimized for AI workflows. A single SN40L chip can hold hundreds of models (trillions of parameters) in memory! The speed alone is a huge deal, but the big advantage is for agentic workflows running multiple specialized models. A GPU can only host a single model and switch (unload and load) to a different model if necessary. An SN40L, on the other hand, can host every model at once, making it much faster. Here is the video where you can see how fast these chips are:

Santiago

97,561 views • 1 year ago

This is the best way to use Claude Fable in Claude Code without immediately hitting your limits. 1. Model set to Fable 5 2. Reasoning on Max 3. Instruct Claude to run a dynamic workflow where: 3a. Fable is the orchestrator 3b. Opus does the reasoning heavy phases Fable is so overpowered that you don't need its intelligence for every step. Let it orchestrate Opus or even Sonnet.

This is the best way to use Claude Fable in Claude Code without immediately hitting your limits. 1. Model set to Fable 5 2. Reasoning on Max 3. Instruct Claude to run a dynamic workflow where: 3a. Fable is the orchestrator 3b. Opus does the reasoning heavy phases Fable is so overpowered that you don't need its intelligence for every step. Let it orchestrate Opus or even Sonnet.

Dan McAteer

911,772 views • 3 months ago

You can now try Llama 3.1 405B for free (link below)! This is the largest open-source model out there, and for the first time, an open model is competitive with closed models. This time around, Meta did something new: Llama 3.1 has a license that allows developers to use it to enhance other models. For the first time, you can distill Llama 3.1 405B's capabilities into a smaller, more practical model for your use case. First, here is the link where you can play with Llama 3.1 for free: The model is hosted in Tune Studio, an end-to-end platform for developing applications using Large Language Models. They are sponsoring this post. Take a look at the attached video. It will show you how you can fine-tune a simple model using Llama 3.1 without leaving the platform: 1. You can create an empty dataset 2. Use the playground to generate and record interactions with Llama 3.1 3. Modify the dataset directly using the playground 4. Export the data and fine-tune a smaller model Fast and easy! As long as you have a web browser, you can start experimenting with fine-tuning and Llama 3.1. That's all it takes!

You can now try Llama 3.1 405B for free (link below)! This is the largest open-source model out there, and for the first time, an open model is competitive with closed models. This time around, Meta did something new: Llama 3.1 has a license that allows developers to use it to enhance other models. For the first time, you can distill Llama 3.1 405B's capabilities into a smaller, more practical model for your use case. First, here is the link where you can play with Llama 3.1 for free: The model is hosted in Tune Studio, an end-to-end platform for developing applications using Large Language Models. They are sponsoring this post. Take a look at the attached video. It will show you how you can fine-tune a simple model using Llama 3.1 without leaving the platform: 1. You can create an empty dataset 2. Use the playground to generate and record interactions with Llama 3.1 3. Modify the dataset directly using the playground 4. Export the data and fine-tune a smaller model Fast and easy! As long as you have a web browser, you can start experimenting with fine-tuning and Llama 3.1. That's all it takes!

Santiago

55,609 views • 2 years ago

Got an output from the strongest pre-release Gemini 3.5 Pro checkpoint 👀 I compared it to Claude Fable 5 and GPT-5.5 on the same exact prompt Yeah... This model is mostly dead on arrival (tried it on other prompts too ofc). It will likely be the "poor man's model" next to 3.5 flash. 1st video: Gemini 3.5 Pro (pre-release) 2nd video: GPT-5.5 3rd video: Claude Fable 5

Got an output from the strongest pre-release Gemini 3.5 Pro checkpoint 👀 I compared it to Claude Fable 5 and GPT-5.5 on the same exact prompt Yeah... This model is mostly dead on arrival (tried it on other prompts too ofc). It will likely be the "poor man's model" next to 3.5 flash. 1st video: Gemini 3.5 Pro (pre-release) 2nd video: GPT-5.5 3rd video: Claude Fable 5

Lentils

50,395 views • 3 months ago

sonnet 5 is routing to sonnet 5.5 it's been a week since it started routing to newer model and team already confirmed that sonnet 5.5 is coming soon from early testing it looks like a huge upgrade from sonnet 5 like opus 5 to opus 5.5 just look at this output, it looks way better then gpt 6 sol i think it's either anthropic is training their models on voxels or sol is just bad please openai don't let sonnet mog sol

sonnet 5 is routing to sonnet 5.5 it's been a week since it started routing to newer model and team already confirmed that sonnet 5.5 is coming soon from early testing it looks like a huge upgrade from sonnet 5 like opus 5 to opus 5.5 just look at this output, it looks way better then gpt 6 sol i think it's either anthropic is training their models on voxels or sol is just bad please openai don't let sonnet mog sol

J A Z I I

21,610 views • 7 days ago

As someone who trains and fine-tunes LLMs, this backpropagation video is the closest thing to a "the calc 1 rule that trains every LLM on earth" explainer I've ever seen released for free. Everyone thinks training GPT-5 requires exotic math. It doesn't. The entire $500B AI industry runs on the chain rule from freshman calculus - the exact rule Marvin Minsky said would never work. Bookmark this 30-min video and watch tonight. Same math your high school teacher taught you, now updating every weight in Claude, GPT-5, and Llama.

As someone who trains and fine-tunes LLMs, this backpropagation video is the closest thing to a "the calc 1 rule that trains every LLM on earth" explainer I've ever seen released for free. Everyone thinks training GPT-5 requires exotic math. It doesn't. The entire $500B AI industry runs on the chain rule from freshman calculus - the exact rule Marvin Minsky said would never work. Bookmark this 30-min video and watch tonight. Same math your high school teacher taught you, now updating every weight in Claude, GPT-5, and Llama.

fintex

50,749 views • 1 month ago

I built a complex history feature for my Figma to Code plugin in 3 prompts. This is a 30-min tutorial using Claude Code. The difference in code generation between gpt-4o, Claude 3.5 and 3.7 is insane. 3.7 produces near-perfect results and is far more consistent. It's really a designer's best friend, not missing details from the design, like adaptive layout, outlines, spacing, etc. It seems like the new model has great taste now.

I built a complex history feature for my Figma to Code plugin in 3 prompts. This is a 30-min tutorial using Claude Code. The difference in code generation between gpt-4o, Claude 3.5 and 3.7 is insane. 3.7 produces near-perfect results and is far more consistent. It's really a designer's best friend, not missing details from the design, like adaptive layout, outlines, spacing, etc. It seems like the new model has great taste now.

Meng To

32,475 views • 1 year ago

Starting today, open source is leading the way. Introducing Llama 3.1: Our most capable models yet. Today we’re releasing a collection of new Llama 3.1 models including our long awaited 405B. These models deliver improved reasoning capabilities, a larger 128K token context window and improved support for 8 languages among other improvements. Llama 3.1 405B rivals leading closed source models on state-of-the-art capabilities across a range of tasks in general knowledge, steerability, math, tool use and multilingual translation. The models are available to download now directly from Meta or Hugging Face. With today’s release the ecosystem is also ready to go with 25+ partners rolling out our latest models — including Amazon Web Services, NVIDIA, Databricks, GroqInc, Dell Technologies, Microsoft Azure and Google Cloud ready on day one. More details in the full announcement ➡️ Download Llama 3.1 models ➡️ With these releases we’re setting the stage for unprecedented new opportunities and we can’t wait to see the innovation our newest models will unlock across all levels of the AI community.

Starting today, open source is leading the way. Introducing Llama 3.1: Our most capable models yet. Today we’re releasing a collection of new Llama 3.1 models including our long awaited 405B. These models deliver improved reasoning capabilities, a larger 128K token context window and improved support for 8 languages among other improvements. Llama 3.1 405B rivals leading closed source models on state-of-the-art capabilities across a range of tasks in general knowledge, steerability, math, tool use and multilingual translation. The models are available to download now directly from Meta or Hugging Face. With today’s release the ecosystem is also ready to go with 25+ partners rolling out our latest models — including Amazon Web Services, NVIDIA, Databricks, GroqInc, Dell Technologies, Microsoft Azure and Google Cloud ready on day one. More details in the full announcement ➡️ Download Llama 3.1 models ➡️ With these releases we’re setting the stage for unprecedented new opportunities and we can’t wait to see the innovation our newest models will unlock across all levels of the AI community.

AI at Meta

1,270,194 views • 2 years ago

Greg Brockman seems to confirm what I’ve suspected for a long time: o1, o3, and GPT-5 Instant/Thinking were all built on the GPT-4o base model People found this hard to believe when I said this 2 years ago, but internally, the original o1 was called GPT-4o with reasoning

Greg Brockman seems to confirm what I’ve suspected for a long time: o1, o3, and GPT-5 Instant/Thinking were all built on the GPT-4o base model People found this hard to believe when I said this 2 years ago, but internally, the original o1 was called GPT-4o with reasoning

Flowers ☾

141,129 views • 29 days ago

Nah, this can’t be real. Same GPT-5.6 Sol. Same task. Same thinking level. I switched from regular Codex to the Claude Code harness, and the result instantly became way better. Claude Code is somehow making every model cook. GPT-5.6 SOL WAS ALREADY TOP TIER!! 😭

Nah, this can’t be real. Same GPT-5.6 Sol. Same task. Same thinking level. I switched from regular Codex to the Claude Code harness, and the result instantly became way better. Claude Code is somehow making every model cook. GPT-5.6 SOL WAS ALREADY TOP TIER!! 😭

OmedTheVibeCoder

127,040 views • 2 months ago

AI is coming for taste and judgment Claude Code's Boris Cherny: "We have this tendency to think....here's a thing the agent can't do yet, this must be a thing it can never do. But the rate of progress is so quick that something like taste, the agent will be able to do quite soon" co-lead cat says you can see the jump in the models directly. The front-end apps Claude builds have gotten visibly nicer between Sonnet 3.5 and Opus 4.7, and you no longer need to handhold the agent.

AI is coming for taste and judgment Claude Code's Boris Cherny: "We have this tendency to think....here's a thing the agent can't do yet, this must be a thing it can never do. But the rate of progress is so quick that something like taste, the agent will be able to do quite soon" co-lead cat says you can see the jump in the models directly. The front-end apps Claude builds have gotten visibly nicer between Sonnet 3.5 and Opus 4.7, and you no longer need to handhold the agent.

Deirdre Bosa

21,115 views • 4 months ago

Claude Sonnet 4.8 is imminent. Most likely dropping today. Sonnet 4.6 is starting to look outdated. Look at the Flappy Bird test on BridgeBench. GPT 5.5 and Claude Opus 4.7 have moved past it. The model that used to be the default for most vibe coders is falling behind. Anthropic needs Sonnet 4.8 to close the gap. The second it drops, we're running it through every BridgeBench test and posting the results live. New Sonnet day. Let's see what Anthropic has.

Claude Sonnet 4.8 is imminent. Most likely dropping today. Sonnet 4.6 is starting to look outdated. Look at the Flappy Bird test on BridgeBench. GPT 5.5 and Claude Opus 4.7 have moved past it. The model that used to be the default for most vibe coders is falling behind. Anthropic needs Sonnet 4.8 to close the gap. The second it drops, we're running it through every BridgeBench test and posting the results live. New Sonnet day. Let's see what Anthropic has.

Bridgebench

16,715 views • 4 months ago