Posts
Elon Musk
@elonmusk · Nov 8, 2025
Haha exactly 😂
Yun-Ta Tsai@yunta_tsai· Nov 8, 2025Designing an inference chip for robots is actually very difficult.
In data centers each chip is bathed in jacuzzi and babysat by nannies. If they died it would be hot swapped by one of their clones.
The fault rate of GPUs in datacenter is actually quite high. Industrial average
1.2K
2K
25.8K
16.7M
Elon Musk
@elonmusk · Nov 8, 2025
Grok 4 (Blazing) Fast with 2M window context
Brian Roemmele@BrianRoemmele· Nov 8, 2025Boom!
Grok 4 Fast now has a huge 2M token context window!
Why is this important?
This context window can allow substantially documents or code bases to be loaded as a single prompt for interrogation and consideration.
With the added bonus of being insanely fast with almost
2.2K
3.1K
23.7K
5.3M
Zeeshan Patel
@zeeshanp_ · Nov 8, 2025
Continual learning is largely a context compression problem. Compressing endless multimodal bitstreams into dense, reusable learning representations is crucial to deploying such systems in the real world. Building a unified learning representation will be necessary towards
Elon Musk
@elonmusk · Nov 8, 2025
🎯
126
33
589
30.6K
Elon Musk
@elonmusk · Nov 8, 2025
Getting there
Dogan Ural@doganuraldesign· Nov 8, 2025I never thought I’d say this, but…
Grok might have reached Midjourney-level aesthetics
Just try it and tell me what you think
3.8K
4.4K
46.8K
10.8M
Elon Musk
@elonmusk · Nov 8, 2025
Made with Grok Imagine
tetsuo@tetsuoai· Nov 8, 2025𝕲𝖗𝖔k 𝕴𝖒𝖆𝖌𝖎𝖓𝖊·˚ ༘ ☾ ⋆｡˚ ☄︎
6.2K
2.7K
29.2K
7.5M
Tom Nash
@iamtomnash · Nov 7, 2025
Elon’s Got the Trillion. To earn it, $TSLA has to hit a $12 trillion market cap, generate $2 trillion in revenue, and out-profit every industrial giant on Earth. No pressure.
Elon Musk
@elonmusk · Nov 8, 2025
😂
224
46
1.7K
59.9K
🇦🇪 HGS
@Sajwani · Nov 8, 2025
Who’s making these memes 🤣🤣
Elon Musk
@elonmusk · Nov 8, 2025
😂
139
48
1.1K
44.8K
Elon Musk
@elonmusk · Nov 8, 2025
Grok
DogeDesigner@cb_doge· Nov 8, 2025BREAKING: Grok just got a major upgrade. Grok-4-Fast got a big boost.
Massive jump in reasoning quality and completion rate.
• Reasoning mode: 77.5% → 94.1%
• Non-reasoning: 77.9% → 97.9%
Grok just got a whole lot smarter.
Great work xAI team.
6K
5.1K
36.8K
5.9M
P_ThA_P
@juanher39282802 · Nov 8, 2025
Also made with grok AI 😂😂. It’s my dog btw
Elon Musk
@elonmusk · Nov 8, 2025
😂
71
23
605
43.9K
Diligent Denizen 🇺🇸
@DiligentDenizen · Nov 8, 2025
Why you doing @IanCarrollShow like that? 👀🤣
Elon Musk
@elonmusk · Nov 8, 2025
😂
58
26
589
45.9K
Alex
@alexfer · Nov 7, 2025
Higher testosterone than the avg man in western society right now
Elon Musk
@elonmusk · Nov 8, 2025
Tragically true
52
34
714
28.8K
@levelsio
@levelsio · Nov 7, 2025
New mood
Elon Musk
@elonmusk · Nov 8, 2025
I think Sydney was trying to say this to the interviewer:
191
248
1.4K
356.6K
Elon Musk
@elonmusk · Nov 8, 2025
Grok Imagine prompt:
Add a sexy vampire boyfriend with big teeth, pointy elf ears and a big mustache.
6.5K
3.1K
22.4K
4.4M
Elon Musk
@elonmusk · Nov 8, 2025
I just used the above prompt on the Grok image below:
Heisenberg@rovvmut_· Nov 7, 2025Holy moly Grok Imagine's image generation is getting so good 🤯
2.8K
367
2.4K
1.6M
Elon Musk
@elonmusk · Nov 8, 2025
Grok Imagine prompt:
She smiles and says “I will always love you”
9.7K
4.3K
40.9K
11.6M
Elon Musk
@elonmusk · Nov 8, 2025
This is actually a major Grok upgrade, but we decided to release it quietly this time.
Gave us time to smooth out the rough edges.
Mario Nawfal@MarioNawfal· Nov 7, 2025GROK JUST GOT A WHOLE LOT SMARTER
xAI quietly patched Grok-4-Fast... and the before/after is wild.
“Reasoning” mode jumped from 77.5% to 94.1% complete responses.
“Non-reasoning”? Up to a staggering 97.9%.
All thanks to one thing: better injected system prompts.
No press
6.5K
6.3K
57.6K
9.6M
Elon Musk
@elonmusk · Nov 8, 2025
Dogan has good taste
Dogan Ural@doganuraldesign· Nov 8, 2025Grok is beautiful
Text-to-image just got a massive upgrade!
Go check it out, you'll be surprised
Grok image + Grok video
4.7K
2.5K
26.6K
7.2M
Yun-Ta Tsai
@yunta_tsai · Nov 8, 2025
“You work at Tesla?”
Elon Musk
@elonmusk · Nov 8, 2025
🔥😂
67
35
818
30.4K
heiner
@HeinrichKuttler · Nov 8, 2025
In this case I was a fan before.
Amir Efrati@amir· Nov 7, 2025Google paid ~$3 billion to hire AI genius Noam Shazeer.
He’s helped Gemini meaningfully since then but also kept posting inflammatory statements that angered colleagues and prompted Google to censor him.
Elon Musk
@elonmusk · Nov 8, 2025
Same
50
28
553
22.4K
Jeff Dean
@JeffDean · Nov 7, 2025
An exciting new approach for doing continual learning, using nested optimization for enhancing long context processing.
Google Research@GoogleResearch· Nov 7, 2025Introducing Nested Learning: A new ML paradigm for continual learning that views models as nested optimization problems to enhance long context processing. Our proof-of-concept model, Hope, shows improved performance in language modeling. Learn more:
Elon Musk
@elonmusk · Nov 8, 2025
Not sure if it’s super new, as we’ve been doing something along these lines at Tesla for a while, but it does make sense.
The single biggest technical challenge of Tesla self-driving AI is context compression of 1.5GB/s of video to ~2kB/s of control outputs using a puny inference computer without making any mistakes.
Super hard to avoid overly lossy compression at any given step.
164
209
2.8K
1.1M