正在加载视频...

视频加载失败

返回首页

timelapse #86 (15 hrs): - got my first OOM on 8xB200 node - defaulting back to grok-code-fast-1, the fastest reliable coding model with by far most intuitive instruction following, combined with grok 4 fast reasoning to plan before i let grok code work its magic - drank 2 large...

elliotarledge's profile picture

Elliot Arledge

38,668 subscribers

102,928 次观看 • 9 个月前 •via X (Twitter)

科学技术
LIVE

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

timelapse #85 (27.5 hrs): - currently cant rely on any other coding models except grok code fast 1 + grok 4 fast (for complex reasoning grok 4 fast is 20 cents for 1M tokens) - wrote qwen3-next trainer entirely from scratch to make it more managable - each piece completely done by grok-code-fast-1 in cursor as it seems to handle this task pretty well without the grok 4 fast reasoning - take on smaller problems and complete them quickly (makes it easier with 400 toks/sec over the api) - got distributed fp8 qwen3-next trainer running at 0.8 seconds per step on 8xH100s (still need to finish checkpoint loading logic) - perfect timing as the fp8 version of qwen3-next drops as im writing this - ill be in LA in 2 days (will visit SF mid way through as well) - 12.5% margarita - steak dinner with family - gained intuition on FlashAttention in very long context settings - caught up w/ Kearm h/eng and Arnie Ramesh
0:22

timelapse #85 (27.5 hrs): - currently cant rely on any other coding models except grok code fast 1 + grok 4 fast (for complex reasoning grok 4 fast is 20 cents for 1M tokens) - wrote qwen3-next trainer entirely from scratch to make it more managable - each piece completely done by grok-code-fast-1 in cursor as it seems to handle this task pretty well without the grok 4 fast reasoning - take on smaller problems and complete them quickly (makes it easier with 400 toks/sec over the api) - got distributed fp8 qwen3-next trainer running at 0.8 seconds per step on 8xH100s (still need to finish checkpoint loading logic) - perfect timing as the fp8 version of qwen3-next drops as im writing this - ill be in LA in 2 days (will visit SF mid way through as well) - 12.5% margarita - steak dinner with family - gained intuition on FlashAttention in very long context settings - caught up w/ Kearm h/eng and Arnie Ramesh

Elliot Arledge

283,820 次观看 • 9 个月前

I took Grok 4 for a spin this weekend to build this game prototype. I used SuperGrok Chat to generate the initial game prototype and then brought it over to Cursor to continue coding with Grok 4 MAX. Grok 4 in Cursor is like a no-nonsense agent. Doesn't speak much, but delivers the goods. There were moments where I was rate-limited or stuck on a bug or two that I had to get other models to help, but otherwise it's a fast, reliable model to work with. This makes me incredibly excited for Grok Code to launch in August!
0:24

I took Grok 4 for a spin this weekend to build this game prototype. I used SuperGrok Chat to generate the initial game prototype and then brought it over to Cursor to continue coding with Grok 4 MAX. Grok 4 in Cursor is like a no-nonsense agent. Doesn't speak much, but delivers the goods. There were moments where I was rate-limited or stuck on a bug or two that I had to get other models to help, but otherwise it's a fast, reliable model to work with. This makes me incredibly excited for Grok Code to launch in August!

Danny Limanseta

20,481,476 次观看 • 1 年前

BREAKING: Grok 4.1 Fast just claimed the #1 spot in programming model usage on OpenRouter. Grok Code Fast 1 holds the #2 position. xAI's Grok has taken both first and second place, showing how strongly users are choosing it over the rest.
0:20

BREAKING: Grok 4.1 Fast just claimed the #1 spot in programming model usage on OpenRouter. Grok Code Fast 1 holds the #2 position. xAI's Grok has taken both first and second place, showing how strongly users are choosing it over the rest.

DogeDesigner

780,767 次观看 • 7 个月前

Grok 4's combined usage captures ~37% of all top-20 token usage this month on OpenRouter, and Grok Code Fast 1 alone accounts for ~29%. 📲 Get Grok 4 on the iPhone App Store to access Ani.
0:29

Grok 4's combined usage captures ~37% of all top-20 token usage this month on OpenRouter, and Grok Code Fast 1 alone accounts for ~29%. 📲 Get Grok 4 on the iPhone App Store to access Ani.

tetsuo

7,508,653 次观看 • 9 个月前

Anyone got any idea how to enable /fast mode on Claude Code It won't do anything, sometimes it finally shows me a login thing after /enable-usage but then I login and I copy the code and I am back here with no fast mode! It did work on my laptop but not on server :/
0:33

Anyone got any idea how to enable /fast mode on Claude Code It won't do anything, sometimes it finally shows me a login thing after /enable-usage but then I login and I copy the code and I am back here with no fast mode! It did work on my laptop but not on server :/

@levelsio

69,131 次观看 • 3 个月前

timelapse #83 (22 hrs): - it was very easy to dive super deep into anything i needed to (this is what i focused on today because not all days are like this) - finding the grok code fast 1 + grok 4 for deep thinking and verification combo to be super useful in cursor. speed was solid - hard to imagine myself spending many more mental clock cycles in a 24 hr period - had to pull out qwen3-next’s gated deltanet + linear attention from bleeding edge hf transformers to begin implementing a multi-gpu fp8 trainer from scratch. this is so damn bleeding edge and i underestimated how much effort this has and will require lol - lots of diet coke and oats - shipped the template which the core chapters of my book will be built on: - all im missing now is flash attention 1/2 mastery (fa2 tmrw), intuition on making topk faster (for arbitrary row length), what i should and shouldnt teach in cutlass/cute, hopper/blackwell gemm kernel mastery (down to fp4) — shoutout to Pranjal for making this easier for me. his blog post is amazing - caught up w/ Mati Roy - im feeling great mentally but not so good physically as im writing this and about to pass out
0:30

timelapse #83 (22 hrs): - it was very easy to dive super deep into anything i needed to (this is what i focused on today because not all days are like this) - finding the grok code fast 1 + grok 4 for deep thinking and verification combo to be super useful in cursor. speed was solid - hard to imagine myself spending many more mental clock cycles in a 24 hr period - had to pull out qwen3-next’s gated deltanet + linear attention from bleeding edge hf transformers to begin implementing a multi-gpu fp8 trainer from scratch. this is so damn bleeding edge and i underestimated how much effort this has and will require lol - lots of diet coke and oats - shipped the template which the core chapters of my book will be built on: - all im missing now is flash attention 1/2 mastery (fa2 tmrw), intuition on making topk faster (for arbitrary row length), what i should and shouldnt teach in cutlass/cute, hopper/blackwell gemm kernel mastery (down to fp4) — shoutout to Pranjal for making this easier for me. his blog post is amazing - caught up w/ Mati Roy - im feeling great mentally but not so good physically as im writing this and about to pass out

Elliot Arledge

2,255,154 次观看 • 9 个月前

Claude Code vs Codex vs Grok Build: Who builds the best Windows 95 OS in HTML? I benchmarked them live on speed, lines of code, file size… and actual usability. Claude: Fastest (384 lines, 14KB) but basic Grok: Took longer… and built a monster (full Explorer, Paint, interactive Grok terminal, browser, Control Panel, dial-up to AOL, shutdown & more) Codex/GPT-5: Didn’t quite make it Out-of-the-box winner? Grok. By a mile. 👑 Full breakdown + results inside 👀
8:10

Claude Code vs Codex vs Grok Build: Who builds the best Windows 95 OS in HTML? I benchmarked them live on speed, lines of code, file size… and actual usability. Claude: Fastest (384 lines, 14KB) but basic Grok: Took longer… and built a monster (full Explorer, Paint, interactive Grok terminal, browser, Control Panel, dial-up to AOL, shutdown & more) Codex/GPT-5: Didn’t quite make it Out-of-the-box winner? Grok. By a mile. 👑 Full breakdown + results inside 👀

Dan

44,508 次观看 • 1 个月前

Grok Code Fast 1 is versatile across the full stack and is particularly strong at TypeScript, Python, Java, Rust, C++, and Go. Using Grok Code Fast 1, Danny Limanseta built the following game in a day.
0:58

Grok Code Fast 1 is versatile across the full stack and is particularly strong at TypeScript, Python, Java, Rust, C++, and Go. Using Grok Code Fast 1, Danny Limanseta built the following game in a day.

xAI

1,075,875 次观看 • 10 个月前

xAI has just released a STRONG model for vibe coding Grok Code Fast 1 is by far the fastest I've tested 🔥 → Usable in Cline (100% open source) → Also Cursor, Copilot, etc. → 160 toks/s (GPT-5 runs at 50 tok/s) → Free for the launch, 6x cheaper than GPT-5 then More below
0:35

xAI has just released a STRONG model for vibe coding Grok Code Fast 1 is by far the fastest I've tested 🔥 → Usable in Cline (100% open source) → Also Cursor, Copilot, etc. → 160 toks/s (GPT-5 runs at 50 tok/s) → Free for the launch, 6x cheaper than GPT-5 then More below

Paul Couvert

144,713 次观看 • 10 个月前

Grok Code just literally cloned the entire Netflix UI with a one shot prompt From the iconic Hero section to the interactive Carousel, all with a single prompt and a few web-sourced image links It’s not just fast, it's the most cost-effective SOTA model out there ➝ Input: Grok $0.20 vs Claude $3 (15x cheaper) ➝ Output: Grok $1.50 vs Claude $15 (10x cheaper) Here’s how you can do it in VS Code with Kilo Code Grok Code matches Claude Sonnet 4 in performance, but is 10-15x more cost-effective All this for less than $1 SOTA power, fractional cost
1:23

Grok Code just literally cloned the entire Netflix UI with a one shot prompt From the iconic Hero section to the interactive Carousel, all with a single prompt and a few web-sourced image links It’s not just fast, it's the most cost-effective SOTA model out there ➝ Input: Grok $0.20 vs Claude $3 (15x cheaper) ➝ Output: Grok $1.50 vs Claude $15 (10x cheaper) Here’s how you can do it in VS Code with Kilo Code Grok Code matches Claude Sonnet 4 in performance, but is 10-15x more cost-effective All this for less than $1 SOTA power, fractional cost

X Freeze

6,972,865 次观看 • 10 个月前

Grok 4.5 is out!!! here is what you need to know : On cursor bench it matched fable 5 high at 83% cheaper, although the ui generation feels similar to gpt 5.5 in a bad way. Important to note, cursor bench was LEAKED to grok 4.5 and may have skewed the results! My initial thoughts are this could be a insanely cheap opus 4.8 level model that runs insanely fast. Will you be using grok build?
1:04

Grok 4.5 is out!!! here is what you need to know : On cursor bench it matched fable 5 high at 83% cheaper, although the ui generation feels similar to gpt 5.5 in a bad way. Important to note, cursor bench was LEAKED to grok 4.5 and may have skewed the results! My initial thoughts are this could be a insanely cheap opus 4.8 level model that runs insanely fast. Will you be using grok build?

Max Blade

18,675 次观看 • 4 天前

"At the Edge of the Galaxy" – Episode 1 I created this video entirely using AI, specifically the Grok Imagine software. I developed the script with Grok . Then I produced it using all the tools available in Grok Grok Imagine , particularly the Agent mode, which saves a tremendous amount of time. I made this video alone, using just a standard computer. I was blown away by the ability to create dialogue that sounds authentic and worlds filled with such lifelike creatures. The consistency of the characters and locations is excellent, and the realism is breathtaking. I don't see what is stopping anyone today from creating a film entirely conceived using AI with Grok Imagine. If you have any questions, please feel free to ask. I will be happy to answer them. If you would like to support my work, the best way to help is to repost the video. Thank you in advance. As it happens, the video is 4 minutes and 20 seconds long (honestly, I didn't do that on purpose)—that ought to bring me good luck, right? Elon Musk
4:20

"At the Edge of the Galaxy" – Episode 1 I created this video entirely using AI, specifically the Grok Imagine software. I developed the script with Grok . Then I produced it using all the tools available in Grok Grok Imagine , particularly the Agent mode, which saves a tremendous amount of time. I made this video alone, using just a standard computer. I was blown away by the ability to create dialogue that sounds authentic and worlds filled with such lifelike creatures. The consistency of the characters and locations is excellent, and the realism is breathtaking. I don't see what is stopping anyone today from creating a film entirely conceived using AI with Grok Imagine. If you have any questions, please feel free to ask. I will be happy to answer them. If you would like to support my work, the best way to help is to repost the video. Thank you in advance. As it happens, the video is 4 minutes and 20 seconds long (honestly, I didn't do that on purpose)—that ought to bring me good luck, right? Elon Musk

Déborah

100,625 次观看 • 6 天前

GROK CODE FAST 1 TAKES #1 ON OPENROUTER, DEVELOPERS PICK REAL PERFORMANCE OVER HYPE Grok Code Fast 1 just locked in the top spot on OpenRouter’s December 2025 monthly leaderboard, and this one actually matters. Real workloads. Real builders choosing what they trust when things break, scale, or need to ship. According to OpenRouter data, Grok Code Fast 1 processed billions of tokens in production, clearing every major competitor. Claude Sonnet 4, GPT-5 variants, Gemini, DeepSeek , all solid models, all left behind. Developers didn’t debate it on X. They just routed traffic. The reasons are practical, not philosophical. Grok Code Fast 1 is fast. Like actually fast. Low latency, massive 2M context window, strong tool calling, and code outputs that don’t collapse the moment you introduce edge cases. What keeps coming up from builders is reliability. Fewer hallucinations. Better memory across long sessions. It doesn’t fight you. It doesn’t nanny you. It just does the work. Yes, the free trial helped drive initial adoption. But free doesn’t explain retention at this scale. People stayed because it worked better for serious tasks, not toy prompts. The bigger signal is this: when you strip away marketing, guardrail theater, and benchmark gaming, developers reward models that prioritize truth, speed, and usefulness. Source: X Freeze, OpenRouter, Grok
0:30

GROK CODE FAST 1 TAKES #1 ON OPENROUTER, DEVELOPERS PICK REAL PERFORMANCE OVER HYPE Grok Code Fast 1 just locked in the top spot on OpenRouter’s December 2025 monthly leaderboard, and this one actually matters. Real workloads. Real builders choosing what they trust when things break, scale, or need to ship. According to OpenRouter data, Grok Code Fast 1 processed billions of tokens in production, clearing every major competitor. Claude Sonnet 4, GPT-5 variants, Gemini, DeepSeek , all solid models, all left behind. Developers didn’t debate it on X. They just routed traffic. The reasons are practical, not philosophical. Grok Code Fast 1 is fast. Like actually fast. Low latency, massive 2M context window, strong tool calling, and code outputs that don’t collapse the moment you introduce edge cases. What keeps coming up from builders is reliability. Fewer hallucinations. Better memory across long sessions. It doesn’t fight you. It doesn’t nanny you. It just does the work. Yes, the free trial helped drive initial adoption. But free doesn’t explain retention at this scale. People stayed because it worked better for serious tasks, not toy prompts. The bigger signal is this: when you strip away marketing, guardrail theater, and benchmark gaming, developers reward models that prioritize truth, speed, and usefulness. Source: X Freeze, OpenRouter, Grok

Mario Nawfal

39,820 次观看 • 6 个月前

Keir Starmer: We need to deal with AI bots. This is something relatively new, and we saw with Grok—did you see the thing with Grok just the other week? An AI bot that allowed you to instruct it to undress people—and it did. It was disgusting, absolutely disgusting. We went into battle on that because I thought this is not good enough. We went into battle, we made the necessary steps, we threatened action, and we won that battle in relation to Grok. But we've got to do it with all AI bots, and we will do it. We can do that by amending some of the legislation that we've already got in place.
0:40

Keir Starmer: We need to deal with AI bots. This is something relatively new, and we saw with Grok—did you see the thing with Grok just the other week? An AI bot that allowed you to instruct it to undress people—and it did. It was disgusting, absolutely disgusting. We went into battle on that because I thought this is not good enough. We went into battle, we made the necessary steps, we threatened action, and we won that battle in relation to Grok. But we've got to do it with all AI bots, and we will do it. We can do that by amending some of the legislation that we've already got in place.

Clash Report

63,266 次观看 • 4 个月前

Grok 4 Heavy is better than any model available at identifying issues in your codebase. Here's the JS prompt I use with my game code to have Grok 4 Heavy find the bugs. Python prompt in Comments👇
1:08

Grok 4 Heavy is better than any model available at identifying issues in your codebase. Here's the JS prompt I use with my game code to have Grok 4 Heavy find the bugs. Python prompt in Comments👇

tetsuo

1,814,530 次观看 • 1 年前

BREAKING 🚨: xAI released "Grok 4 Fast" in Early beta on the Grok web. Here is a comparison of Grok 4 vs. Grok 4 Fast in terms of speed. Fastest AI on the Earth? 👀
0:36

BREAKING 🚨: xAI released "Grok 4 Fast" in Early beta on the Grok web. Here is a comparison of Grok 4 vs. Grok 4 Fast in terms of speed. Fastest AI on the Earth? 👀

TestingCatalog News 🗞

75,101 次观看 • 10 个月前

I got to try Grok 4.5 in early access in Cursor for the past few days and I absolutely enjoyed it. It feels like Opus 4.8 at 2x the speed at a much cheaper price point. I tasked it to brainstorm > plan > implement a big feature for my game (this act 1 boss fight) and it did not disappoint. - It is much smarter than Composer 2.5, during planning mode, it is able to think through my request more robustly, ensuring that edge cases are covered and makes sure to ask the right questions to confirm with me first. - It is much better at brainstorming ideas/suggestions, similar to Opus 4.8, though I think Fable still edges out a little when it comes to brainstorming ideas and suggestions - It is FAST. probably the fastest of all frontier models (Opus 4.8, GPT 5.5 etc), which makes it a joy to build with, because I can stay in the flow - It has much improved visual/animation capabilities than Composer 2.5, it can code up animations (i wanted an explosion animation with particle effects) with much, much better visuals, animation movement and timing. This is a big leap and I was so happy to see this improvement. - The best part for me is that I can just use the same model from planning down to execution without switching to a lower cost model because the price point is cheaper than other frontier models. I'll be testing this model with more challenging tasks in the next few days but I think this is going to be my main driver for vibe coding for a while. Also, its nice to see Grok back in the race. 🙌
0:34

I got to try Grok 4.5 in early access in Cursor for the past few days and I absolutely enjoyed it. It feels like Opus 4.8 at 2x the speed at a much cheaper price point. I tasked it to brainstorm > plan > implement a big feature for my game (this act 1 boss fight) and it did not disappoint. - It is much smarter than Composer 2.5, during planning mode, it is able to think through my request more robustly, ensuring that edge cases are covered and makes sure to ask the right questions to confirm with me first. - It is much better at brainstorming ideas/suggestions, similar to Opus 4.8, though I think Fable still edges out a little when it comes to brainstorming ideas and suggestions - It is FAST. probably the fastest of all frontier models (Opus 4.8, GPT 5.5 etc), which makes it a joy to build with, because I can stay in the flow - It has much improved visual/animation capabilities than Composer 2.5, it can code up animations (i wanted an explosion animation with particle effects) with much, much better visuals, animation movement and timing. This is a big leap and I was so happy to see this improvement. - The best part for me is that I can just use the same model from planning down to execution without switching to a lower cost model because the price point is cheaper than other frontier models. I'll be testing this model with more challenging tasks in the next few days but I think this is going to be my main driver for vibe coding for a while. Also, its nice to see Grok back in the race. 🙌

Danny Limanseta

1,311,553 次观看 • 4 天前

grok 3 is so awesome, i can't get enough of it. the level of excitement is on par with how i felt when i tried ChatGPT's first public release but grok is WAY better. i only know how to code myspace pages (thanks, tom) and earlier i created this game. what do you think?
0:52

grok 3 is so awesome, i can't get enough of it. the level of excitement is on par with how i felt when i tried ChatGPT's first public release but grok is WAY better. i only know how to code myspace pages (thanks, tom) and earlier i created this game. what do you think?

Hodler

2,447,777 次观看 • 1 年前

🇺🇸 TOM GREEN: ROGAN WAS FIRST TO NAIL THE STREAMING BUSINESS MODEL "In the early 2000s, I realized you could stream live on the internet. I’d been experimenting with tech since the '90s—my website launched in 1994, with audio clips posted even back then. When the technology caught up, I built a multi-camera TV studio in my LA living room and started streaming. I didn’t figure out the business model—Joe Rogan nailed that—but I figured out how to plug it all in. Rogan was way ahead when it came to making it work." Source: Graham Bensinger Tom Green
2:27

🇺🇸 TOM GREEN: ROGAN WAS FIRST TO NAIL THE STREAMING BUSINESS MODEL "In the early 2000s, I realized you could stream live on the internet. I’d been experimenting with tech since the '90s—my website launched in 1994, with audio clips posted even back then. When the technology caught up, I built a multi-camera TV studio in my LA living room and started streaming. I didn’t figure out the business model—Joe Rogan nailed that—but I figured out how to plug it all in. Rogan was way ahead when it came to making it work." Source: Graham Bensinger Tom Green

Mario Nawfal

231,073 次观看 • 1 年前

100% Grok coding It might depend on the language, but I got the impression that the code output volume at once is the richest✍️ The coding itself, along with its input materials and output destinations, all feel like they're part of Elon Musk 's evolving circular activities. The full code is posted in the reply⬇️
0:32

100% Grok coding It might depend on the language, but I got the impression that the code output volume at once is the richest✍️ The coding itself, along with its input materials and output destinations, all feel like they're part of Elon Musk 's evolving circular activities. The full code is posted in the reply⬇️

Yohei Nishitsuji

52,249 次观看 • 10 个月前