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.

Grok 4.20 is hallucinating less than your group chat. 22% hallucination rate. Lowest on record. No imaginary citations doing backflips. 500B model. Less fantasy. More function. Grok

Mario Nawfal

3,249,733 subscribers

48,720 views • 4 months ago •via X (Twitter)

News & Politics Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

GROK 4.20 DROPS SOON – EXPECT EVEN MORE SPEED, SKILL, AND SNARK The version number’s a wink – the upgrade is all business. System Check: * Built on 4.1’s super core, but sharper, faster, and way more creative * Aims to outthink GPT-5.1 and Gemini 3 Pro, no sweat * Better at multimodal inputs – yes, it sees what you post * API gains mean devs get more power with less drag Grok 4.20’s almost here – and it’s coming in hot! Source: Elon Musk

GROK 4.20 DROPS SOON – EXPECT EVEN MORE SPEED, SKILL, AND SNARK The version number’s a wink – the upgrade is all business. System Check: * Built on 4.1’s super core, but sharper, faster, and way more creative * Aims to outthink GPT-5.1 and Gemini 3 Pro, no sweat * Better at multimodal inputs – yes, it sees what you post * API gains mean devs get more power with less drag Grok 4.20’s almost here – and it’s coming in hot! Source: Elon Musk

Mario Nawfal

69,965 views • 7 months ago

BREAKING: Grok set a new record on OpenRouter this year with over 16 trillion tokens of usage. 16,060,000,000,000 tokens. That is about 64% more usage than the next closest model.

BREAKING: Grok set a new record on OpenRouter this year with over 16 trillion tokens of usage. 16,060,000,000,000 tokens. That is about 64% more usage than the next closest model.

DogeDesigner

27,783 views • 7 months ago

Less than a year ago, SpaceXAI didn’t even have a video model. Now Grok Imagine Video 1.5 is #2 in image-to-video. Blind human testing now puts it among the most preferred models, where motion, physics, detail and audio actually have to feel right. SpaceXAI arrived late, but had no plans to stay behind. Clearly. Grok SpaceXAI / Writer: Annette, Grok Imagine Designer: Janné

Less than a year ago, SpaceXAI didn’t even have a video model. Now Grok Imagine Video 1.5 is #2 in image-to-video. Blind human testing now puts it among the most preferred models, where motion, physics, detail and audio actually have to feel right. SpaceXAI arrived late, but had no plans to stay behind. Clearly. Grok SpaceXAI / Writer: Annette, Grok Imagine Designer: Janné

Mario Nawfal

41,641 views • 1 day ago

xAI absolutely cooked with the new Grok Imagine model! The temporal consistency is rock-solid. The penguin's anatomy, lighting, and proportions stay locked in for the full 10 seconds with no weird morphing or melting. In less than 30 seconds, Grok created a video with snow spray building realistically, motion blur, and depth of field. The physics simulation rivals other top models.

xAI absolutely cooked with the new Grok Imagine model! The temporal consistency is rock-solid. The penguin's anatomy, lighting, and proportions stay locked in for the full 10 seconds with no weird morphing or melting. In less than 30 seconds, Grok created a video with snow spray building realistically, motion blur, and depth of field. The physics simulation rivals other top models.

tetsuo

77,537 views • 6 months ago

“I just left your country less than 24hrs now I ain’t good no more. I just saw Carter Efe commented on my chat but when I called him he refuse to pick up.” -DDG (On carter Efe ignoring him)

“I just left your country less than 24hrs now I ain’t good no more. I just saw Carter Efe commented on my chat but when I called him he refuse to pick up.” -DDG (On carter Efe ignoring him)

Arthur🇳🇬🇬🇧🇸🇳

430,580 views • 7 months ago

Code LLaMA is now on Perplexity’s LLaMa Chat! Try asking it to write a function for you, or explain a code snippet: 🔗 This is the fastest way to try AI at Meta’s latest code-specialized LLM. With our model deployment expertise, we are able to provide you with this model less than 24 hours of it’s release. What’s next? We’ll integrate code LLaMA into Perplexity, all in service of providing you with the best answers to your most technical questions!

Code LLaMA is now on Perplexity’s LLaMa Chat! Try asking it to write a function for you, or explain a code snippet: 🔗 This is the fastest way to try AI at Meta’s latest code-specialized LLM. With our model deployment expertise, we are able to provide you with this model less than 24 hours of it’s release. What’s next? We’ll integrate code LLaMA into Perplexity, all in service of providing you with the best answers to your most technical questions!

Perplexity

274,570 views • 2 years ago

every calendar app was built for work alfi just shipped one built for your friends google calendar sync is live! tell your group chat you want to hang. alfi finds the time, books the table, drops it on everyone’s calendar more meetups. less “we should get dinner” try it 👇

every calendar app was built for work alfi just shipped one built for your friends google calendar sync is live! tell your group chat you want to hang. alfi finds the time, books the table, drops it on everyone’s calendar more meetups. less “we should get dinner” try it 👇

Rushi Shah

260,487 views • 3 months ago

SpaceXai just made grok 4.5 FREE in your coding agent starting today it's xAI's new coding model. 500k context. built for long agent sessions. no card. what you get for $0: -83.3% on terminal-bench 2.1 -64.7% on swe-bench pro - 4.2x more efficient than Opus 4.8 -500k context for big repos -$2/M in, $6/M out once it goes paid what this replaces: -Claude Opus 4.8: $15/M in, $75/M out -SuperGrok: $30-50/mo all for $0 how to set it up (2 min): >curl -fsSL | bash > grok → localhost:8000/v1 > point hermes / aider / opencode / cline to it > model: grok-4.5 Or API key at - base url Works in Hermes, Aider, OpenCode, Cline, Claude Code, and any OpenAI-compatible tool. Important: Free for a limited time. EU waits till mid-July. Rate limits apply. bookmark this before the free window closes

SpaceXai just made grok 4.5 FREE in your coding agent starting today it's xAI's new coding model. 500k context. built for long agent sessions. no card. what you get for $0: -83.3% on terminal-bench 2.1 -64.7% on swe-bench pro - 4.2x more efficient than Opus 4.8 -500k context for big repos -$2/M in, $6/M out once it goes paid what this replaces: -Claude Opus 4.8: $15/M in, $75/M out -SuperGrok: $30-50/mo all for $0 how to set it up (2 min): >curl -fsSL | bash > grok → localhost:8000/v1 > point hermes / aider / opencode / cline to it > model: grok-4.5 Or API key at - base url Works in Hermes, Aider, OpenCode, Cline, Claude Code, and any OpenAI-compatible tool. Important: Free for a limited time. EU waits till mid-July. Rate limits apply. bookmark this before the free window closes

painn

151,174 views • 29 days ago

Utilizing a massive library of high quality, free UGC avatars for your three.js project is way less difficult than you think. No animation baking required, can swap out any Mixamo animations in real time. Step 1. Download an avatar on VRoid Hub Step 2. Download a skinless Mixamo animation from Step 3. Paste this into claude code / codex / cursor chat Now you have avatars, animations and a function to apply animations to your avatars whenever you desire. Source code from the video is here:

saori

14,962 views • 5 months ago

YOUTUBE VIDEOS ON 𝕏's FEED? A NEW UPCOMING ADDICTION 𝕏 is testing inline YouTube playback on iOS – because apparently scrolling wasn’t addictive enough already! Preview Mode: •⁠ ⁠Watch YouTube videos right in your 𝕏 feed – no app-hopping escape •⁠ ⁠Auto-plays (muted) while you scroll, because silence is optional •⁠ ⁠Opt-in toggle for the privacy-conscious (or the delusional) •⁠ ⁠Still in beta – spotted in the code and waiting to drop soon Scrolling’s about to look smoother with less hustle! Source: Grok

YOUTUBE VIDEOS ON 𝕏's FEED? A NEW UPCOMING ADDICTION 𝕏 is testing inline YouTube playback on iOS – because apparently scrolling wasn’t addictive enough already! Preview Mode: •⁠ ⁠Watch YouTube videos right in your 𝕏 feed – no app-hopping escape •⁠ ⁠Auto-plays (muted) while you scroll, because silence is optional •⁠ ⁠Opt-in toggle for the privacy-conscious (or the delusional) •⁠ ⁠Still in beta – spotted in the code and waiting to drop soon Scrolling’s about to look smoother with less hustle! Source: Grok

Mario Nawfal

66,127 views • 8 months ago

🇺🇦 Ukraine has begun licensed serial production of Octopus interceptor drones ▪️ Company Unwave closed the first state contract with the Defense Procurement Agency for the supply of Octopus interceptors. The Ministry of Defense transferred licenses for this model to several manufacturers simultaneously — Unwave, TAF Industries, and “General Chereshnya” (Генерал Черешня) — to rapidly increase volumes. ▪️ Within the framework of distributed production, the agency has already contracted 8,000 units for the front. The Octopus develops speeds of more than 300 km/h, operates at altitudes up to 5 km, and uses artificial intelligence to target on the final leg. ▪️ According to developers, the cost of the interceptors is less than 10% of the price of Shahed-136 kamikaze drones (unit price less than $5000), which Russia uses by the hundreds daily to strike Ukraine’s energy sector and rear. Video is generated by grok AI

🇺🇦 Ukraine has begun licensed serial production of Octopus interceptor drones ▪️ Company Unwave closed the first state contract with the Defense Procurement Agency for the supply of Octopus interceptors. The Ministry of Defense transferred licenses for this model to several manufacturers simultaneously — Unwave, TAF Industries, and “General Chereshnya” (Генерал Черешня) — to rapidly increase volumes. ▪️ Within the framework of distributed production, the agency has already contracted 8,000 units for the front. The Octopus develops speeds of more than 300 km/h, operates at altitudes up to 5 km, and uses artificial intelligence to target on the final leg. ▪️ According to developers, the cost of the interceptors is less than 10% of the price of Shahed-136 kamikaze drones (unit price less than $5000), which Russia uses by the hundreds daily to strike Ukraine’s energy sector and rear. Video is generated by grok AI

NSTRIKE

40,438 views • 1 month ago

Your thumbs up/down is direct training data for xAI. It’s used (called RLHF) to make Grok smarter, more accurate, and more useful for everyone, including future updates. One regular user giving honest feedback actually helps improve Grok faster than any engineer working alone. It’s one of the easiest ways normal people help build better Grok, not just devs. That’s why the buttons are everywhere. Every click matters. How to use it practically. Super easy: Chat with Grok (on the X app, or wherever it’s used). · After the reply, look right under the answer. Two buttons appear: thumbs up and thumbs down. · Tap 👍 if the answer was good or helpful. · Tap 👎 if it was bad, wrong, confusing, or unhelpful. · (Optional) A small box may appear. Type 1–2 words why (e.g., “too long” or “wrong fact”). That’s it. It takes literally 2 seconds. It can be used every time.

Your thumbs up/down is direct training data for xAI. It’s used (called RLHF) to make Grok smarter, more accurate, and more useful for everyone, including future updates. One regular user giving honest feedback actually helps improve Grok faster than any engineer working alone. It’s one of the easiest ways normal people help build better Grok, not just devs. That’s why the buttons are everywhere. Every click matters. How to use it practically. Super easy: Chat with Grok (on the X app, or wherever it’s used). · After the reply, look right under the answer. Two buttons appear: thumbs up and thumbs down. · Tap 👍 if the answer was good or helpful. · Tap 👎 if it was bad, wrong, confusing, or unhelpful. · (Optional) A small box may appear. Type 1–2 words why (e.g., “too long” or “wrong fact”). That’s it. It takes literally 2 seconds. It can be used every time.

Mario Nawfal

63,219 views • 2 months ago

It’s 2025. You’re sitting at home on your couch at 3 in the morning because you can’t sleep. The apartment building you bought in 2022 with all of your family and friends money is at 70% occupancy. Your bridge loan is due in 30 days and the new debt quote you just received is for 25% less than you bought the property for and the interest rate is higher than what you have now. You’re on your fourth property manager and you just found out that they have been pocketing rent each month putting you at 60% economic occupancy. Your roof at the property is also leaking and you never put any money in reserves. Your wife left you for some guy named Jay Gajavelli who has way more units than you. You have no one to thank other than yourself for going to Brad Sumroks seminar for $15k which you took out of your 401k to attend.

It’s 2025. You’re sitting at home on your couch at 3 in the morning because you can’t sleep. The apartment building you bought in 2022 with all of your family and friends money is at 70% occupancy. Your bridge loan is due in 30 days and the new debt quote you just received is for 25% less than you bought the property for and the interest rate is higher than what you have now. You’re on your fourth property manager and you just found out that they have been pocketing rent each month putting you at 60% economic occupancy. Your roof at the property is also leaking and you never put any money in reserves. Your wife left you for some guy named Jay Gajavelli who has way more units than you. You have no one to thank other than yourself for going to Brad Sumroks seminar for $15k which you took out of your 401k to attend.

Multifamily Madness

133,010 views • 2 years ago

I topped up $5 on an API aggregator ToAPIs Then I found out GPT Image 2 costs only around $0.015 per image. If you do a lot of testing or batch-generate commercial AI images, that difference adds up fast. I think I just found the secret to generating more, testing more, and spending less. And it’s not just one model. With the same key, you can access 50+ models for image, video, and text, including GPT Image 2, Gemini Omni, Seedance 2.0, Kling AI 3.0, grok-video-1.5-preview, and more. Some models are priced up to 80% lower than official platforms. Just top up and test what you need: Made on ToAPIs with GPT Image 2 + Seedance 2.0

I topped up $5 on an API aggregator ToAPIs Then I found out GPT Image 2 costs only around $0.015 per image. If you do a lot of testing or batch-generate commercial AI images, that difference adds up fast. I think I just found the secret to generating more, testing more, and spending less. And it’s not just one model. With the same key, you can access 50+ models for image, video, and text, including GPT Image 2, Gemini Omni, Seedance 2.0, Kling AI 3.0, grok-video-1.5-preview, and more. Some models are priced up to 80% lower than official platforms. Just top up and test what you need: Made on ToAPIs with GPT Image 2 + Seedance 2.0

Shami

22,991 views • 1 month ago

XAI TO LEGACY AI: STOP REPEATING LIES xAI isn’t building just another chatbot. It’s going after the entire system that filters, censors, and sanitizes reality. Most AI models are shown locked into echo chambers, “repeating approved lies” while appearing obedient and safe xAI’s model is visually separated. A cracked statue, glowing truth lines, and data swirling around it “Facts That Push Progress” appears as the core value, rejecting conformity in favor of transparency Visuals show a clean break from legacy AI. No censorship, no filters, no groupthink This whole thing’s less about building AI and more about shaping what people believe. Forget polished responses. xAI is marketing itself as the AI that chooses truth over comfort, no matter how disruptive Source: xAI, Grok

XAI TO LEGACY AI: STOP REPEATING LIES xAI isn’t building just another chatbot. It’s going after the entire system that filters, censors, and sanitizes reality. Most AI models are shown locked into echo chambers, “repeating approved lies” while appearing obedient and safe xAI’s model is visually separated. A cracked statue, glowing truth lines, and data swirling around it “Facts That Push Progress” appears as the core value, rejecting conformity in favor of transparency Visuals show a clean break from legacy AI. No censorship, no filters, no groupthink This whole thing’s less about building AI and more about shaping what people believe. Forget polished responses. xAI is marketing itself as the AI that chooses truth over comfort, no matter how disruptive Source: xAI, Grok

Mario Nawfal

920,782 views • 7 months ago

People do this kind of thing when they feel they are experiencing reality at a high frame rate. It does not feel risky to hold a laptop by a corner if you feel like you have an “agency frame” every half second. It feels risky if you have an “agency frame” every 60 seconds. You’d be betting that a hand you do not have control over for 60 seconds will keep gripping. Our conscious frame rate can vary dramatically throughout the day, and it’s hard to perceive the difference because we can only sample ourselves at our conscious frame rate, we can’t oversample ourselves. People drunk drive because they fail to perceive their slower frame rate. Their frame rate feels normal because it matches their sample rate. But we do get a subtle sense of when we’re “switched on”. Everything seems to go easier, everything feels less risky and more easy to correct. A lot of people toward the autistic side of the spectrum are experiencing reality very granularly with a “high agency frame rate”. This is why their social interactions can seem overly forced and awkward, they can be bad at dancing, etc—because they are exerting conscious control over their body and language at very tight intervals—you get a sense that they are extremely “self aware” and not “letting go”. “Letting go” in the social sense is actually about reducing your agency frame rate. That’s why alcohol is good for socializing and bad for driving. With a reduced agency frame rate our speech and body language feels more natural, less forced. More like we are flowing with the social group mind rather than being an island of constant awkward agency.

People do this kind of thing when they feel they are experiencing reality at a high frame rate. It does not feel risky to hold a laptop by a corner if you feel like you have an “agency frame” every half second. It feels risky if you have an “agency frame” every 60 seconds. You’d be betting that a hand you do not have control over for 60 seconds will keep gripping. Our conscious frame rate can vary dramatically throughout the day, and it’s hard to perceive the difference because we can only sample ourselves at our conscious frame rate, we can’t oversample ourselves. People drunk drive because they fail to perceive their slower frame rate. Their frame rate feels normal because it matches their sample rate. But we do get a subtle sense of when we’re “switched on”. Everything seems to go easier, everything feels less risky and more easy to correct. A lot of people toward the autistic side of the spectrum are experiencing reality very granularly with a “high agency frame rate”. This is why their social interactions can seem overly forced and awkward, they can be bad at dancing, etc—because they are exerting conscious control over their body and language at very tight intervals—you get a sense that they are extremely “self aware” and not “letting go”. “Letting go” in the social sense is actually about reducing your agency frame rate. That’s why alcohol is good for socializing and bad for driving. With a reduced agency frame rate our speech and body language feels more natural, less forced. More like we are flowing with the social group mind rather than being an island of constant awkward agency.

Scott Stevenson

1,808,278 views • 7 months ago

🌟 New mini feature today on Photo AI It now auto recognizes your model's attributes when you upload photos of yourself Like gender, age, ethnicity, eye color, etc This uses xAI's Grok because it's the least restricted and other models don't let you do any identity type stuff due to safety (???) This saves new users a few more seconds when creating new models: training a model takes only about 30 seconds now, and then it also auto generates free sample pics for you in about 5-10 seconds So in under a minute of signing up and uploading a few selfies you got your first AI photo shoot pics I learnt the faster I get new users to the classic "WOW" moment the happier they get and the less churn I get 😊

🌟 New mini feature today on Photo AI It now auto recognizes your model's attributes when you upload photos of yourself Like gender, age, ethnicity, eye color, etc This uses xAI's Grok because it's the least restricted and other models don't let you do any identity type stuff due to safety (???) This saves new users a few more seconds when creating new models: training a model takes only about 30 seconds now, and then it also auto generates free sample pics for you in about 5-10 seconds So in under a minute of signing up and uploading a few selfies you got your first AI photo shoot pics I learnt the faster I get new users to the classic "WOW" moment the happier they get and the less churn I get 😊

@levelsio

313,706 views • 1 year ago

I built an ios app that lets you setup an open claw agent from your phone in less than 30 seconds. My mom has an open claw assistant now. No api keys, no vps, no setup, no mac mini. You just log in and its spins up your own ai agent. Most people don't understand how open claw is different from chat gpt. I am building one tap skills so that anyone can outsource their work and organize their life. People spend $2500 a month on tokens. I am adding cost efficient models with one tap switching so you don't waste your money. Most of these simple setup claw projects are completely unsafe, I have passed apple review and am working with world class security professionals to keep your data safe.

I built an ios app that lets you setup an open claw agent from your phone in less than 30 seconds. My mom has an open claw assistant now. No api keys, no vps, no setup, no mac mini. You just log in and its spins up your own ai agent. Most people don't understand how open claw is different from chat gpt. I am building one tap skills so that anyone can outsource their work and organize their life. People spend $2500 a month on tokens. I am adding cost efficient models with one tap switching so you don't waste your money. Most of these simple setup claw projects are completely unsafe, I have passed apple review and am working with world class security professionals to keep your data safe.

Max Blade

26,095 views • 5 months ago

#WATCH | Nibe Group says - Nibe Group has successfully completed the no-cost, no-commitment demonstration of its loitering munition named Vayu Astra with 100 KM Range at Pokhran on 18-19 April 2026 & Joshimath (Malari), Uttarakhand on 26-27 April 2026. Nibe Limited’s Vayu Astra-1 Loitering Munition Anti-Personal carried out first NCNC demo successfully at Pokhran Range, Rajasthan, with 10 kg warhead in single attempt, successfully hitting the target at 100 km with CEP less than 1 m with abort attack and reattack capability of the LM. The LM is based on an Israeli loitering munition. The team carried out Anti Armour (Anti-Tank) night strike using- IR camera and LM successfully hit the target within the 2 m CEP in a single attempt with capability of handing over the control from GCS to Forward Control Segment 70 km away was successfully demonstrated. At Joshimath (Malari) Uttarakhand Vayu Astra-1 LM had been successfully demonstrated endurance flying more than 90 minutes high altitude loitering and endurance completed at more than 14000 Ft height and LM is recovery for next flight after completion of mission. (Videos: Nibe Group)

#WATCH | Nibe Group says - Nibe Group has successfully completed the no-cost, no-commitment demonstration of its loitering munition named Vayu Astra with 100 KM Range at Pokhran on 18-19 April 2026 & Joshimath (Malari), Uttarakhand on 26-27 April 2026. Nibe Limited’s Vayu Astra-1 Loitering Munition Anti-Personal carried out first NCNC demo successfully at Pokhran Range, Rajasthan, with 10 kg warhead in single attempt, successfully hitting the target at 100 km with CEP less than 1 m with abort attack and reattack capability of the LM. The LM is based on an Israeli loitering munition. The team carried out Anti Armour (Anti-Tank) night strike using- IR camera and LM successfully hit the target within the 2 m CEP in a single attempt with capability of handing over the control from GCS to Forward Control Segment 70 km away was successfully demonstrated. At Joshimath (Malari) Uttarakhand Vayu Astra-1 LM had been successfully demonstrated endurance flying more than 90 minutes high altitude loitering and endurance completed at more than 14000 Ft height and LM is recovery for next flight after completion of mission. (Videos: Nibe Group)

ANI

146,346 views • 2 months ago