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。

Grok 4.7 vs SWE 2 tested both models with same prompt at highest reasoning available and results came out really different > grok 4.7 took 32 minutes and cost $8.14 to make this > swe 2 took 15 minutes and cost $0 to make this really surprised by how... show more

J A Z I I

9,925 subscribers

50,932 次观看 • 1 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

43 条评论

J A Z I I 的头像

J A Z I I1 天前

grok was tested in grok build at xhigh reasoning and swe was tested in devin desktop app at max reasoning was expecting more from grok but well it is what it is prompt: create a beautiful, busy, living koi pond using Three.js in a single HTML file. Include a waterfall, rocks, plants, and multiple koi fish with different colors and patterns. Make the scene feel alive through continuous fish movement, flowing water, and animated environmental details. Add a “Feed the koi” button. When clicked, food should appear on the water, and the fish should visibly swim toward it and eat it—not simply continue their normal swimming animation. Allow the camera to orbit and zoom so I can explore the pond. Make the scene visually polished, with convincing water, lighting, and depth. Deliver the complete working HTML file.

Bee 的头像

Bee1 天前

SWE 2 is so good wtf.. It's not even comparable with Grok 4.7

J A Z I I 的头像

J A Z I I1 天前

wait for frontend test

Bee 的头像

Bee1 天前

Waiting.. could you try this too? Here's the prompt..

J A Z I I 的头像

J A Z I I1 天前

lemme do it

Dave Alexander 的头像

Dave Alexander1 天前

swe 2 is not free

J A Z I I 的头像

J A Z I I1 天前

free in all subs

Fanfulla 的头像

Fanfulla1 天前

If you think that SWE-2 is free right now is kinda insane

FeiZ 的头像

FeiZ1 天前

Oh no, grok 4.7 looks bad.

J A Z I I 的头像

J A Z I I1 天前

yeah just worst

Hakm 的头像

Hakm1 天前

SWE >>

Lumina 的头像

Lumina1 天前

are you using Grok 4.7 inside Grok build though, it's ass outside of it, like unusable

J A Z I I 的头像

J A Z I I1 天前

yes ser

Lumina 的头像

Lumina1 天前

thats awful then 💀

Vjekoslav Jug 的头像

Vjekoslav Jug1 天前

After trying it out I have to say Grok 4.7 results feels kinda weird

Zephyr ✪ 的头像

Zephyr ✪1 天前

swe 2 is underrated

Yohaku 的头像

Yohaku1 天前

of course i prefer for Grok 4.7 here

remakw 的头像

remakw1 天前

unfortunately quite tough task for grok

J A Z I I 的头像

J A Z I I1 天前

yeah should have given more harder task

Fakhr 的头像

Fakhr1 天前

swe 2… just wow 😮 Why is it so underrated….

Shubh 的头像

Shubh1 天前

swe 2 actually did a great job chief 👀

Haleemah 的头像

Haleemah1 天前

Swe is way better but grok did good too

avgrocketman 的头像

avgrocketman1 天前

grok 4.7 is so bad lmao

Knowix 的头像

Knowix1 天前

they both came out great, but I think SWE 2 performed better here

KC 的头像

KC1 天前

SWE 2

Hybrid 🏃🏾 的头像

Hybrid 🏃🏾1 天前

Bro, SWE-2 blew grok out of the water. Pun intended

J A Z I I 的头像

J A Z I I1 天前

haha swe the goat

Hybrid 🏃🏾 的头像

Hybrid 🏃🏾1 天前

I mean it really is. @cognition cooked

EVOKER 的头像

EVOKER1 天前

Awsome

balega_dev 的头像

balega_dev1 天前

I swear saw this exact video before just with different models 😂

MJ 的头像

MJ1 天前

SWE 2 🎉

Eloi 的头像

Eloi1 天前

I do not think this is a fair comparison SWE-2 is free until October and max may use a crazy amount of token that make it more expensive as well.

lisham 的头像

lisham1 天前

man, it's so obvious. 4.7 doesn't come any closer to swe 2. truth has to be said.

Guilherme Figueiredo 的头像

Guilherme Figueiredo1 天前

Try to make an engineer test on both of them, I'm pretty sure grok will win.

Zach | Engineer → AI builder 的头像

Zach | Engineer → AI builder1 天前

I get SWE 2 is free right now but that is not always going to be the case. It did cook here though.

Digital Dweller 的头像

Digital Dweller1 天前

So about $60 billion down the drain. Cursor was supposed to turn it around for them but Grok is still cooked by open source models.

Aight Man 的头像

Aight Man1 天前

Dead on fucking arrival lmao.

Marcel 的头像

Marcel1 天前

i think it’s a tie i haven’t tried SWE 2 yet, it looks so good. where can i try it out?

Jakub Arnold 的头像

Jakub Arnold1 天前

swe-2 is not free though?

luojiuyue 的头像

luojiuyue1 天前

wow！what's the SWE2？the style looks like kimi k3！

Ryan Sael 的头像

Ryan Sael1 天前

swe 2 looks interesting is this right inside devin or other harness?

陈晨的头像

陈晨1 天前

0美元不可能，如果你说他在devin会员中免费，那么任何一家ai公司的前沿模型都是免费的，包括Astra，fable5.1

Decatalyst 🌱 的头像

Decatalyst 🌱1 天前

How many parameters is Swe ?

相关视频

GPT 6 Astra vs SWE 2.0 tested both models with same prompt at highest reasoning available > astra took 80 minutes to complete this task and costed $64 > swe 2.0 took 25 minutes and costed zero dollars really surprised by swe results here and astra looks nerfed right now, no idea why which one did better here?

GPT 6 Astra vs SWE 2.0 tested both models with same prompt at highest reasoning available > astra took 80 minutes to complete this task and costed $64 > swe 2.0 took 25 minutes and costed zero dollars really surprised by swe results here and astra looks nerfed right now, no idea why which one did better here?

J A Z I I

31,322 次观看 • 12 天前

Opus 5 vs DeepSeek v4.1 Flash tested both models with same frontend prompt at highest reasoning available but results came out really different > opus took 80 minutes to complete and costed $20 > v4.1 flash took 60 minutes and costed $0.2 which one did better here?

Opus 5 vs DeepSeek v4.1 Flash tested both models with same frontend prompt at highest reasoning available but results came out really different > opus took 80 minutes to complete and costed $20 > v4.1 flash took 60 minutes and costed $0.2 which one did better here?

J A Z I I

39,045 次观看 • 12 天前

GLM 5.3 Flash vs Kimi K3 tested both models on same prompt at highest reasoning available but results came out really different > 5.3 flash took 20 minutes to complete this and costed $0.7 via AIhubMix API > kimi k3 took 30 minutes to complete this and costed $4.32 via Kimi Plan which one do you think did better here?

GLM 5.3 Flash vs Kimi K3 tested both models on same prompt at highest reasoning available but results came out really different > 5.3 flash took 20 minutes to complete this and costed $0.7 via AIhubMix API > kimi k3 took 30 minutes to complete this and costed $4.32 via Kimi Plan which one do you think did better here?

J A Z I I

23,105 次观看 • 27 天前

Fable 5.1 vs Kimi K3 Tested both models with the same prompt at the highest reasoning level available. > Fable took 18 minutes to complete this test and cost $12.60 > Kimi K3 took 10 minutes and cost around $6.95 Which one wins here?

Fable 5.1 vs Kimi K3 Tested both models with the same prompt at the highest reasoning level available. > Fable took 18 minutes to complete this test and cost $12.60 > Kimi K3 took 10 minutes and cost around $6.95 Which one wins here?

Bhavy☄️

147,705 次观看 • 21 天前

deepseek v4.1 flash vs secret model tested both with the same prompt at the highest reasoning available > v4.1 flash took 2 hours to finish both tests and cost $2.60 > secret model finished both in 30 minutes, can’t reveal cost yet v4.1 flash spent a lot of time overthinking and running random tests, which slowed it down thanks AIHubmix for providing api access so i could test v4.1 flash hoping that team will cut down the overthinking before the full release, but for now here are the results which model did better here?

deepseek v4.1 flash vs secret model tested both with the same prompt at the highest reasoning available > v4.1 flash took 2 hours to finish both tests and cost $2.60 > secret model finished both in 30 minutes, can’t reveal cost yet v4.1 flash spent a lot of time overthinking and running random tests, which slowed it down thanks AIHubmix for providing api access so i could test v4.1 flash hoping that team will cut down the overthinking before the full release, but for now here are the results which model did better here?

J A Z I I

28,763 次观看 • 14 天前

Grok 4.7 just released and it's an EXCELLENT model It was trained FOR Grok Bot Meaning this is a fully agentic model trained to do your knowledge work better than you can In this video I show you how to use Grok 4.7 and a Grok Bot workflow that will 10x your productivity:

Grok 4.7 just released and it's an EXCELLENT model It was trained FOR Grok Bot Meaning this is a fully agentic model trained to do your knowledge work better than you can In this video I show you how to use Grok 4.7 and a Grok Bot workflow that will 10x your productivity:

Alex Finn

500,685 次观看 • 1 天前

meta muse just mogged fable 5.1 tested fable 5.1 and muse spark 1.3 with same prompt at highest reasoning available and results came out really different > muse spark 1.3 completed task in one minute and costed almost nothing > fable 5.1 completed task in 70 minutes and costed $13 it's clear that meta is stepping up their game in ai, they don't wanna left behind which one did better here?

meta muse just mogged fable 5.1 tested fable 5.1 and muse spark 1.3 with same prompt at highest reasoning available and results came out really different > muse spark 1.3 completed task in one minute and costed almost nothing > fable 5.1 completed task in 70 minutes and costed $13 it's clear that meta is stepping up their game in ai, they don't wanna left behind which one did better here?

J A Z I I

75,566 次观看 • 20 天前

DeepSeek V4 Flash 0731 in Hermes Agent and one prompt, took 32 minutes and cost 0.07$, this model is so cheap to the point where 2 dollars can last you a full day.

DeepSeek V4 Flash 0731 in Hermes Agent and one prompt, took 32 minutes and cost 0.07$, this model is so cheap to the point where 2 dollars can last you a full day.

Elshayib

689,761 次观看 • 1 个月前

@Grok is no longer just a chatbot. 35 cents. 2 prompts. A fully functional 3D holographic SpaceX Starship simulator. Grok Build 0.1 did in minutes what used to cost thousands of dollars and weeks of engineering hours.

@Grok is no longer just a chatbot. 35 cents. 2 prompts. A fully functional 3D holographic SpaceX Starship simulator. Grok Build 0.1 did in minutes what used to cost thousands of dollars and weeks of engineering hours.

Mario Nawfal

84,008 次观看 • 3 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 $1 SOTA power, fractional cost$

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

6,973,710 次观看 • 1 年前

Opus 4.8 vs MiniMax M3 tested both on default settings with the same prompt > Opus one shotted everything in 7 minutes > M3 needed an extra prompt to fix the "break block" feature and took 20+ minutes both got super close, judge both and lemme know which one looks better?

Opus 4.8 vs MiniMax M3 tested both on default settings with the same prompt > Opus one shotted everything in 7 minutes > M3 needed an extra prompt to fix the "break block" feature and took 20+ minutes both got super close, judge both and lemme know which one looks better?

J A Z I I

67,862 次观看 • 3 个月前

ELON MUSK: Grok 5 will train on all SpaceX data, which could make it “by far the best engineer”. “We’re making rapid progress on Grok because Grok 4.5 was a huge improvement. And we have Grok 4.6 coming out probably next week. And then Grok 4.7 is about three or four weeks from today. And we expect the cadence of AI development to improve dramatically. “And then, with Grok 5, which should be out before the end of this year, we will be incorporating the entire corpus of SpaceX data. Basically, all the data that SpaceX has ever produced, which is a tremendous amount over the course of a quarter century, will be incorporated into Grok training. And we think this will make Grok by far the best engineer.”

ELON MUSK: Grok 5 will train on all SpaceX data, which could make it “by far the best engineer”. “We’re making rapid progress on Grok because Grok 4.5 was a huge improvement. And we have Grok 4.6 coming out probably next week. And then Grok 4.7 is about three or four weeks from today. And we expect the cadence of AI development to improve dramatically. “And then, with Grok 5, which should be out before the end of this year, we will be incorporating the entire corpus of SpaceX data. Basically, all the data that SpaceX has ever produced, which is a tremendous amount over the course of a quarter century, will be incorporated into Grok training. And we think this will make Grok by far the best engineer.”

DogeDesigner

190,101 次观看 • 1 个月前

Fable 5 vs GPT 5.5 tested both on High settings with the same prompt > Fable one shotted everything in 15 minutes and used 3% of weekly limits > GPT 5.5 took 20 minutes and used 3% of weekly limits 5.5 is not looking good here at all which one do you think won?

Fable 5 vs GPT 5.5 tested both on High settings with the same prompt > Fable one shotted everything in 15 minutes and used 3% of weekly limits > GPT 5.5 took 20 minutes and used 3% of weekly limits 5.5 is not looking good here at all which one do you think won?

J A Z I I

107,777 次观看 • 3 个月前

Look how much Grok has progressed in just 6 months from Grok 3 to Grok 4.6! As a quick Saturday morning project I gave Grok Build the link to this previous "𝕏 THE GAME" viral (yet ugly) game Grok was able to watch the video, quickly reverse engineer the game, and make dramatic improvements You can play it here (on desktop): Some lessons I learned: - Grok is extremely token efficient. I was able to iterate on the design many times and still kept the token cost surprisingly low (only a few $ to make this game) - Grok is great at working with humans. Instead of just telling it what to do, or on the other extreme having it go off and do something itself, ask for a list of options on how to improve and tell Grok which ones to implement. A true partnership - Anyone can build things these days with AI, no code knowledge needed What are you building with Grok?

Look how much Grok has progressed in just 6 months from Grok 3 to Grok 4.6! As a quick Saturday morning project I gave Grok Build the link to this previous "𝕏 THE GAME" viral (yet ugly) game Grok was able to watch the video, quickly reverse engineer the game, and make dramatic improvements You can play it here (on desktop): Some lessons I learned: - Grok is extremely token efficient. I was able to iterate on the design many times and still kept the token cost surprisingly low (only a few $ to make this game) - Grok is great at working with humans. Instead of just telling it what to do, or on the other extreme having it go off and do something itself, ask for a list of options on how to improve and tell Grok which ones to implement. A true partnership - Anyone can build things these days with AI, no code knowledge needed What are you building with Grok?

Dan

6,151,532 次观看 • 1 个月前

I just created this 36-second video in just five minutes with a simple prompt using Agent Mode in Grok Imagine All it took was a simple prompt, and it was made in just five minutes With a single image, it can generate matching scenes from different angles, build a full visual storyline, and even add subtitles. Grok Imagine Agent is one of the best tools for visual storytelling It's amazing how quickly you can now bring an idea to life just by expressing it in a few words The barrier between imagination and creation is getting smaller every day

I just created this 36-second video in just five minutes with a simple prompt using Agent Mode in Grok Imagine All it took was a simple prompt, and it was made in just five minutes With a single image, it can generate matching scenes from different angles, build a full visual storyline, and even add subtitles. Grok Imagine Agent is one of the best tools for visual storytelling It's amazing how quickly you can now bring an idea to life just by expressing it in a few words The barrier between imagination and creation is getting smaller every day

X Freeze

26,471 次观看 • 2 个月前

Just built this space battle game with Grok 4.5 in Grok Build, and it absolutely nailed it I literally gave it a simple two-line prompt, asked it to research the idea and build the game, and it did the rest A few back-and-forth tweaks, and took a free low-poly spaceship model, and Grok handled the integration surprisingly well....even the sound effects Now I can’t stop playing it. It’s actually super fun. Grok just got even more addictive…

Just built this space battle game with Grok 4.5 in Grok Build, and it absolutely nailed it I literally gave it a simple two-line prompt, asked it to research the idea and build the game, and it did the rest A few back-and-forth tweaks, and took a free low-poly spaceship model, and Grok handled the integration surprisingly well....even the sound effects Now I can’t stop playing it. It’s actually super fun. Grok just got even more addictive…

X Freeze

100,978 次观看 • 2 个月前

Elon Musk: “We’re making rapid progress on Grok because Grok 4.5 was a huge improvement We have Grok 4.6 coming out probably next week, and then Grok 4.7 is about three or four weeks from today. We expect the cadence of AI development to improve dramatically And then, with Grok 5....which should be out before the end of the year.....we will be incorporating the entire corpus of SpaceX: basically all the data that SpaceX has ever produced That is a tremendous amount over the course of a quarter century, and it will be incorporated into Grok training We think this will make Grok, by far, the best engineer…”

Elon Musk: “We’re making rapid progress on Grok because Grok 4.5 was a huge improvement We have Grok 4.6 coming out probably next week, and then Grok 4.7 is about three or four weeks from today. We expect the cadence of AI development to improve dramatically And then, with Grok 5....which should be out before the end of the year.....we will be incorporating the entire corpus of SpaceX: basically all the data that SpaceX has ever produced That is a tremendous amount over the course of a quarter century, and it will be incorporated into Grok training We think this will make Grok, by far, the best engineer…”

X Freeze

39,005 次观看 • 1 个月前

Just tested Kimi.ai K3 against GPT-5.6 and Grok 4.5 on the same task: build a Subway Surfers game. Kimi K3 was my favorite. Better UI, smoother gameplay, and the overall experience felt more polished. GPT-5.6 was solid too, while Grok 4.5 fell behind on the UI. The interesting part is that all three needed at least one follow-up prompt. Kimi also took a bit longer to generate, but the extra time paid off. Ran Kimi K3 through Nebius Token Factory. Low inference cost + a really capable open source model is a great combo.

Just tested Kimi.ai K3 against GPT-5.6 and Grok 4.5 on the same task: build a Subway Surfers game. Kimi K3 was my favorite. Better UI, smoother gameplay, and the overall experience felt more polished. GPT-5.6 was solid too, while Grok 4.5 fell behind on the UI. The interesting part is that all three needed at least one follow-up prompt. Kimi also took a bit longer to generate, but the extra time paid off. Ran Kimi K3 through Nebius Token Factory. Low inference cost + a really capable open source model is a great combo.

Arindam Majumder 𝕏

46,107 次观看 • 1 个月前

fable 5.2 > gpt 6 astra tested both models with same prompt at high reasoning and results came out really different fable next model entered stealth testing today, so i don’t think it's coming out anytime soon from early testing: > huge step up from current fable > output looks better than astra > slower and expensive still running more tests and will drop the outputs soon

fable 5.2 > gpt 6 astra tested both models with same prompt at high reasoning and results came out really different fable next model entered stealth testing today, so i don’t think it's coming out anytime soon from early testing: > huge step up from current fable > output looks better than astra > slower and expensive still running more tests and will drop the outputs soon

J A Z I I

39,450 次观看 • 4 天前

Genuinely impressed by Grok 4.6 Test it with different prompts. Put it next to other models. The results keep coming out strong. And this was just one simple prompt: Create a 3d Queen Anne's Revenge shipusing three.js in white themebackground so ship is properly visible

Genuinely impressed by Grok 4.6 Test it with different prompts. Put it next to other models. The results keep coming out strong. And this was just one simple prompt: Create a 3d Queen Anne's Revenge shipusing three.js in white themebackground so ship is properly visible

Shawn

145,989 次观看 • 1 个月前