Ryan Brewer

1.9K posts

Ryan Brewer banner
Ryan Brewer

Ryan Brewer

@ryanbrewer

building for finance @OpenAI

NY | SF Katılım Aralık 2018
1.3K Takip Edilen8K Takipçiler
Sabitlenmiş Tweet
Ryan Brewer
Ryan Brewer@ryanbrewer·
Excited to share that I’ve joined OpenAI to help build the future of financial intelligence. My focus is connecting models directly to the data sources, tools, and workflows analysts use every day. My team’s first product is the Excel plugin. After software engineering, finance will see the benefits of model improvements more acutely than almost any other field. This is just the start for us! More to come soon!
OpenAI@OpenAI

GPT-5.4 Thinking and GPT-5.4 Pro are rolling out now in ChatGPT. GPT-5.4 is also now available in the API and Codex. GPT-5.4 brings our advances in reasoning, coding, and agentic workflows into one frontier model.

English
59
47
990
347.9K
Ryan Brewer retweetledi
Gabriel Chua
Gabriel Chua@gabrielchua·
Preparing for tomorrow’s demo—just wanted to pause and appreciate how seamlessly spreadsheets and slide decks are presented in ChatGPT Work on the web. And yes: the numbers are entirely fictional for the demo :)
Gabriel Chua tweet media
English
3
4
35
2.7K
Ryan Brewer
Ryan Brewer@ryanbrewer·
Touched grass
Ryan Brewer tweet mediaRyan Brewer tweet mediaRyan Brewer tweet mediaRyan Brewer tweet media
English
3
0
26
1.3K
Ryan Brewer
Ryan Brewer@ryanbrewer·
@mweinbach 0 need to be at a desk other than to collaborate with teammates
English
1
0
4
633
Max Weinbach
Max Weinbach@mweinbach·
My view is people use AI at work to improve efficiency and quality of life. If I can delegate from my phone to an AI to do what I’d normally do while at my desk, why do I need to be at my desk? More productive and better quality of life and work life balance
Sam Altman@sama

so far at least, i'm pretty sure AI has been net job-creating. this was not what i expected--although i was much less pessimistic than others, i thought by this level of capability we'd have seen some impact. it is possible this direction keeps going!

English
10
10
250
25.1K
David Cramer
David Cramer@zeeg·
@ryanbrewer This is our review harness - you can mostly ignore the Pi detail, it was more important when we were evaluating the cost of Claude SDK vs Pi's SDK
English
1
0
1
402
David Cramer
David Cramer@zeeg·
Alright I got one more benchmark for the weekend, but I dont think it changes my opinion right now that Grok 4.5 might be best bang for the buck GPT 5.6 Luna on High on Warden's security bench. Competitive-enough on pricing and good accuracy. Added the precision cost to the table to help normalize it a bit.
David Cramer tweet media
English
12
6
81
9K
Ryan Brewer
Ryan Brewer@ryanbrewer·
@iMuffined Intentional! Model will continue churning if you’re in the middle of the turn
English
2
1
8
285
iMuffin
iMuffin@iMuffined·
I hit my 5h limit multiple times but it just keeps working through it?
iMuffin tweet media
English
8
0
31
1.4K
jason
jason@jxnlco·
Rage baiting for views is so silly to me cause you’re betraying your own values for what, 120$ of Twitter payouts?
English
65
9
411
22.3K
Ryan Brewer
Ryan Brewer@ryanbrewer·
Thank god this was my biggest gripe with gpt models
Arena.ai@arena

Exciting news: @OpenAI’s GPT-5.6-sol is now joint #1 in the Code Arena: Frontend, matching Claude Fable 5! This marks the first time an OpenAI model has reached the top spot in Code Arena, demonstrating major gains in agentic coding, frontend and web app development. Highlights: - Significant improvement from GPT-5.5-xhigh (#18 -> #1) - #1 in Data & Analytics, Brand Marketing, Consumer product, and Gaming - Priced at $5/$30 per million input/output tokens - roughly 2× cheaper than Claude Fable 5 Huge congrats to the @OpenAI team for this incredible milestone!

English
0
1
64
5.6K
Ryan Brewer
Ryan Brewer@ryanbrewer·
@kcosr Fwiw I do every day without unlimited resets
English
0
0
2
98
Kevin
Kevin@kcosr·
@ryanbrewer It will go faster than it did 2 days ago! You guys really should have to dog food using subs (with unlimited resets, of course) so that you can see how harness changes (e.g. subagent attributes) will impact users. 😃
English
1
0
1
164
Ryan Brewer
Ryan Brewer@ryanbrewer·
I'm smurfing on a personal pro account let's see how fast I can use my 5 hour limit with sol ultra
English
6
0
24
2.3K
Ryan Brewer
Ryan Brewer@ryanbrewer·
RT @jxnlco: re: 5.6 sol post trained luna and we won't stop you from doing the same thing To be clear, this was not a model inventing a w…
English
0
4
0
117
Ryan Brewer retweetledi
Neel Ajjarapu
Neel Ajjarapu@neelajj·
GPT-5.6 creates much better natively-editable slides, sheets, and docs! So impressed with how far we've come in a short amount of time
Neel Ajjarapu tweet media
English
10
3
94
6.9K
rohit
rohit@krishnanrohit·
So Codex app is now chatgpt? Is chatgpt also codex? But chatgpt dot com has chat, which doesn't have Sol, only 5.5. It has Work, which has both. Codex also has Work. But it's now called chatgpt codex.
English
18
2
83
7.7K
Ryan Brewer retweetledi
Datacurve
Datacurve@datacurve·
GPT-5.6 tops the DeepSWE leaderboard at 73%. Sol, Terra, and Luna results are now available.
English
70
170
2K
322.4K

Keşfet

@mweinbach @zeeg @iMuffined @angadsg @jxnlco @kcosr @ycjgt @elonmusk