One of the best things I've told my agents is this
"Re Read the plan, ask any questions where intent is not clarified. Once intent is clarified implement the plan and don't stop to ask questions, you own all architectural decisions, I own intent only."
1
1
9
Astra 6.2 is the #1 model on our multi-turn coding eval: Vibe Code Bench 1→100.
Other benchmarks test a model’s ability to handle a single prompt. This benchmark, created by Vals, tests the ability to modify an evolving codebase over many user requests while preserving existing functionality without introducing regressions.
Ben retweeted
We’re going to see something common during anthropics ipo, all big ai influencers who bad mouthed gpt 6.2 beyond what it deserved will buy anthropic hard. Bad reviews, differentiated opinions are fine but calling it unusable, refusing to post a full review is total bs. I mean if you look at their posts, you can smell the crap. Everybody is out to make some money.
AND If they truly mean what they say for the right reasons, choose better words next time guys
OPENAI is under attack and i am on OpenAIs side, which side are you on?
Ben retweeted
I tested GPT-6.2 Astra on real work. 2 repos, 105 replanted bugs. Find and fix what you can.
It's the new #1. The results:
- GPT-6.2 Astra (max): 62.7 for $28.35
- GPT-6.2 Astra (xhigh): 60 for $23.53
- Fable 5.2 (max): 58 for $139.65
- Gemini 4 Carbon (high): 57.3 for $31.32
- Sonnet 5.5 (max): 51.3 for $154.03
- GPT-6 Astra (max): 45 for $33.03
From GPT-6 Astra to GPT-6.2 Astra: +17.7 points at max, +17 at xhigh.
Whatever people say about other use cases, it's a code review beast.
GPT-6.2 Sol and more effort levels drop soon in the thread 🧵
Ben retweeted
Two blind code reviews, 26 real bugs — Sol 6.2 found more than Opus 5.5. That's the sentence that matters, not the bench, the real repo. If the cheap one reviews like this, the expensive one has a problem. Still, small sample. I want ten.
I ran two quick code review tests based on real-world work in @Kullevi_com's codebase, comparing GPT-6.2 Sol against GPT-6.1 Sol, Opus 5.5, and Fable 5.2.
Based on these tests, at least, 6.2 seems to hold its own.
Ben retweeted
I almost forgot my favourite GPT-6.2-Astra generation - this is SVG (yes SVG!!) of a peacock - no other model came close to the precision, lifelike nature and detail of this generation, incl. Fable
Opus 5.5 is a model in its own right, it is probably the best model ever released, and the reason is was such an amazing model was every model since 4.5 was complete crap, until 5.5. It blew our minds away. OpenAI have been making consistently good models since 5.3 probably even earlier, we aren't going to get blown away by them as much because they already do such an amazing job.
1
31
Ran ChatGPT Astra 6.2 across the World Famous Alex Finn Benchmark Test ™
It scored a 312.7 out of 500. Claude Fable 5.2 beat it by .3
But Astra 6.2 finished the benchmark in a third of the time for a third of the price at a quarter of the tokens
The panicking influencers never seem to take into account efficiency and price
This is a great model
Fable orchastrating using Sol 6.2 as sub agents, pretty sure this model isn't what the "influences" are saying it is.
@thsottiaux @OpenAI @OpenAIDevs
29
I don't undestand the hate on 6.2 Sol and Astra. I've been running it flat out all day along side Opus and Fable and its every bit as good at agentic coding. Its picked up things Claude missed, Claude has picked up things Codex missed, that always happens, but where I find it better is its faster, talks less and has concrete evidence to back up its claims not "I though the file existed because I read it in a stale doc at some point in time".
Seems like a highly capable model. Guess I don't get an Anthropic shares.
@OpenAI @thsottiaux @sama @OpenAIDevs
1
1
45
Just like how the light bulb was invented, someone left a candle on overnight.
Theo claims LLMs were discovered when an autocomplete training run went way too long. Like someone left autocomplete on over the weekend and came back to ChatGPT. LLMs exist bc of the transformer and nobody found that by accident. What the fuck is he talking about
1
72
Needs some more testing but I feel the code base direction of this fork is in much better shape than the original branch.
T3 Code v0.1.7: hand-written TS files over 800 lines from 361 to 0. 65 commits, zero behaviour changes. Clean typechecks, tests byte-identical to baseline.
Written by open-weight DeepSeek V4.1 Flash on an office box. A human drove it via Telegram and wrote none of it.
@t3dotcodes
github.com/pingdotgg/t3code/…
1
2
58
Ben retweeted
Theos 1300$ AI course has 0 graduates and 5 stars. The review is from a different course. Right under it sits an OpenAI logo and the site never says why. In the code the whole section is called "Social proof"
Ben retweeted
Theo told you not to buy his 1300$ AI course. He is right.👍
They sell it as Joel Hooks nervously pitching a guy who doesnt do courses. The public repo of Joels company shows them building devrel dot fyi in November 2023, the paid DevRel course Theo announced that fall. Joel pushed fixes himself, they archived it a year later, and the page still collects emails.
MEGA is built by BRAVE, a polish course company that calls its american partner Joel Higgs😭 on its own site. The architect is a guy named Adam, and the site says his team develops the core lesson content while the four stars add insights. The repo students get has sixteen commits, most of it for Theos week, all sixteen from Adam. Theo only shows up in it as a made up character in the sample data.
Angie Jones left the teleprompter script of her first lesson in her public repo for a day. Every sentence written out with stage cues, target 30 minutes, the script runs 20. Twenty lessons at that length is under ten hours of video for 1300$.
Under the "you should not buy it" post Theo dropped a link for your boss. It ends in ref=theo and the sites own code calls that affiliate capture, cookie lasts a week. The same code carries a last call popup, a "price goes up in" timer and discounts for poorer countries. No boss needs any of that. That same night he wrote that if the course bombs he has to let go of some of his 18 employees. MEGA went from two thousand buyers to three thousand in ten days.
Theo only rented out his face for a quick BAG
Reading all the comments from @nvidia employees about the leadership of @JensenHuang brings me back to the book that is the foundation of my leadership.
Empowering people to go great things starts with humbling yourself and realising you are part of the team, you are not the team. 12 Elements of Great Managing is one of the best books I have ever read on the topic. We have previously broken each element into a month of the year and lived it to ensure we capture what our amazing people bring to our organisations.
4
57
























