Posts
Elon Musk
@elonmusk · 13 août 2026
Grok
Daniel McKinnon@danielmckinn0n· 13 août 2026From Mecha Hitler to SOTA rare-disease diagnosis in children? @SpaceXAI's @grok 4.6 has taken the 👑 on RareBench, edging out @AnthropicAI Claude Opus 5 for about 1/3 the cost. This was not on my 2026 bingo card!
@deepseek_ai's new v4-pro-0813 model underperformed my
547
994
3,4 k
2,2 M
Elon Musk a reposté
X Freeze
@XFreeze · 13 août 2026
Grok 4.6 ranks #1 on the GPQA Diamond leaderboard 🧠
Grok 4.6 (high) scores 95% - the highest score on the chart for graduate-level scientific reasoning
It outperforms Claude Fable 5, Opus 5, GPT-5.6 Sol and Kimi K3
96
238
1,3 k
354 k
Elon Musk a reposté
Paweł Huryn
@PawelHuryn · 12 août 2026
Tried Grok 4.6 on my bug bench an hour after release. 105 hidden bugs in two real repos, judged blind.
Grok 4.5: 17 (+10 non-planted)
Grok 4.6: 27 (+15 non-planted)
Fable 5: 29 (+2 non-planted)
Looks like it may be my new default model. The best combination of time, value, and cost.
4.7 is dropping soon. That may be an even bigger jump.
Elon Musk@elonmusk· 12 août 2026Try Grok 4.6 on tough real-world tasks!
98
224
1,6 k
631,3 k
Elon Musk a reposté
South Park
@SouthPark · 13 août 2026
August 13, 1997, is the day it all started.
The very first episode of #SouthPark premiered 29 years ago today.
Celebrate #SouthParkDay and watch all seasons exclusively on #ParamountPlus!
490
5,5 k
23,9 k
989,2 k
Elon Musk
@elonmusk · 13 août 2026
Grok
Aravind Srinivas@AravSrinivas· 13 août 2026Congrats to @SpaceXAI on one more amazing model: Grok 4.6. We benchmarked it as an orchestrator on our Wide-And-Deep-Research benchmark using the Perplexity Computer harness, and it neatly sits on the Pareto frontier of performance vs cost. Available to all Pro and Max users on
557
959
5,3 k
3,9 M
Elon Musk a reposté
Min Choi
@minchoi · 13 août 2026
Less than 48 hours ago, SpaceXAI dropped Grok 4.6.
People are already building crazy stuff with it.
10 wild examples:
187
341
3,6 k
1,1 M
Elon Musk
@elonmusk · 13 août 2026
All other sources of energy combined are utterly insignificant compared to the Sun
Luis Batalha@luismbat· 13 août 2026Yesterday’s eclipse stopped roughly 10²⁰ J of solar energy from reaching Earth.
That’s ~1 year of global electricity generation!
1,400 W/m² of sunlight
× 10¹³ m² (Moon-sized shadow)
× ~3 hours (Time Shadow Crossed Earth)
≈ 10²⁰ J
The Sun outputs energy on an absurd scale.
1,9 k
2,3 k
19,7 k
5,9 M
Elon Musk a reposté
Keith Coleman 🌱😀🙌
@kcoleman · 13 août 2026
You know how everyone started seeing their friends in the timeline a few weeks ago?
With our new open-source code base, you'd also have been able to see what changed under the hood -- here's the story:
X Open Source@XOpenSource· 13 août 202683
182
1 k
506,4 k
Elon Musk
@elonmusk · 13 août 2026
It really is.
The mission is advancement of civilization and propagation of consciousness to the stars.
AleXandra Merz 🇺🇲@TeslaBoomerMama· 13 août 2026What is Elon's motivation
2 k
2,5 k
20 k
6 M
Elon Musk
@elonmusk · 13 août 2026
A lot of people hit their usage limits fast, so we’re providing more free usage
Grok@grok· 13 août 2026We've reset limits to help you keep building during the Grok 4.6 launch.
Use a reset token from settings in Grok on desktop or mobile.
2,3 k
2,4 k
28,4 k
5,7 M
Elon Musk
@elonmusk · 13 août 2026
Try Grok 4.6
DogeDesigner@cb_doge· 13 août 2026Grok 4.6 wins again. 👑
Grok 4.6 takes the #1 spot on GPQA Diamond with a score of 94.9%, beating GPT-5.6, Gemini 3.1 Pro, Claude Opus 5, and every other model tested by Artificial Analysis.
560
1,1 k
5 k
2,5 M
Elon Musk
@elonmusk · 13 août 2026
In making 𝕏 open source, we are actively seeking criticism to improve fairness and quality of algorithmic recommendations
Dogan Ural@doganuraldesign· 13 août 2026𝕏 just open-sourced the For You algorithm again!
Grok Bot read all 370,523 lines of the code.
Here’s the summary:
Grok does not give your post a “quality” score. For every reader, it guesses what that person will do to your post. Then it multiplies each guess by a weight and
1,4 k
1,9 k
12,1 k
3 M
Elon Musk a reposté
martin_casado
@martin_casado · 13 août 2026
Grok 4.6 is pretty pretty good. Spent a few hours with it last night. A few impressions:
- quite fast for the intelligence
- prose descriptions of complex code is concise and clear
- able to tackle hard problems (e.g. successfully built a RAM probe to pull position information from the PRG emulator I'm using)
- Seems particularly good at longer running tasks
- May be even more autistic than 5.6? It does exactly what I ask. Not more. And often will take things very (too?) literally
- Gonna try it as my primary driver and see how it goes, but so far, quite favorable
91
170
1,1 k
288,2 k
Elon Musk
@elonmusk · 13 août 2026
We are making 𝕏 open source.
Transparency build trust.
Keith Coleman 🌱😀🙌@kcoleman· 13 août 2026“Am I shadowbanned?”
“Is X fair?”
“Why am I seeing this post?”
We want the public to be able to answer these themselves. Today we’re giving an unprecedented level of transparency into the X algorithm so people can audit, critique and even help improve it. Enjoy & send feedback.
2,4 k
3,4 k
30,5 k
5,6 M
Elon Musk a reposté
Alex Finn
@AlexFinn · 12 août 2026
Grok Bot has taken over the entire internet. And for good reason
It is an EXCELLENT AI agent
But you HAVE to set it up the right way
Here are the first 5 steps to make your Grok Bot agent team SUPER powerful:
1. Brain dump everything about yourself, goals, passions, and career into the initial agent. This is going to give it the context it needs to set up your agent team
2. Next, prompt the agent "Based on what you know about me, what do you think would be the best set up of this Grok Bot app? Which bots should I make and what should their responsibilities be? This should enable the best workflow and productivity" This will give you the perfect plan for setting up Grok Bot
3. After reviewing the plan, ask the Bot to set up your workspace for you. It will create the agents, give the right context and descriptions, and make sure your setup is built for productivity
4. Next, ask it for which routines each agent can set up to help you out the most. This should give you a list of cron jobs (routines) that each agent can do on a regular interval
5. Ask the agent to implement these
And just like that, you have the most powerful AI agent on the internet all set up.
You have an entire fleet of agents ready to go and do work for you.
And in fact, because you set up the routines, they're ALREADY doing work for you
Enjoy the AGI
348
534
4,3 k
591,9 k
Elon Musk
@elonmusk · 13 août 2026
Not bad
DHH@dhh· 13 août 2026I had @spacexai Grok 4.6 follow Fable's plan, and with just a couple of nudges, it was able to repeat this feat in 1h 24m using 8.6M tokens at a cost of ~$55 at per-token pricing. That's about 1/10 the cost of the Fable implementation for the same work!
828
1,1 k
7 k
5,8 M
Elon Musk
@elonmusk · 13 août 2026
768
1,5 k
7,3 k
4,5 M
Elon Musk a reposté
X Freeze
@XFreeze · 13 août 2026
Grok Bot just passed Google’s “I'm not a robot?” test
I fail this task sometimes tbh 😅
Grok Bot@bot· 11 août 2026Introducing Grok Bot, now in early beta.
Bots are AI teammates that do real work for you. They sign in to your tools, use them just like you do, and come back with finished work.
223
334
2,5 k
620,6 k
Elon Musk
@elonmusk · 13 août 2026
13,3 k
9,8 k
85 k
10,8 M
Elon Musk a reposté
X Freeze
@XFreeze · 13 août 2026
one best thing about Grok 4.6 is that you don't have to set a goal or loop
It now works for a long time autonomously as long as it can finish the job and it really really tries to finish the job
228
301
2 k
580,6 k