@sandersted

Research at OpenAI. Be kind to others, and yourself.

 San Francisco, CA
tedsanders.com
Joined September 2009
  • Tweets1.1K
  • Following1.6K
  • Followers11.2K
15Media
Ted Sanders retweeted
Matt Workman
@mattworkman1h
GPT 6.1 Terra (Ultra) is INSANELY good at 3D Blender computer use *only used 3% of my Pro 200 plan > Codex + fal plugin > generate an image of the roadster > Blender 3D model the car and animate the camera move > GPT 2.5 CG first frame to photo real > Seedance 2.5 ref to video 3D video to photo real video
OpenAI
@OpenAI4h
GPT-6.1 Terra: delivering fast and capable computer use at a lower price. GPT-6.1 Terra understands what’s on screen, takes action across apps, writes naturally, and is a more capable model for your whole workday with the efficiency to take on more of it.
10
10
72
7,175
Ted Sanders retweeted
Michael Wall
@sound4movement2h
One of my most favorite model releases to date. I used Terra, it had o3 vibes. 6.1-Terra is on a whole other level and improves on that high-quality rigor. From super creative work to making email actually fun, it’s a great model sir.
OpenAI
@OpenAI4h
GPT-6.1 Terra: delivering fast and capable computer use at a lower price. GPT-6.1 Terra understands what’s on screen, takes action across apps, writes naturally, and is a more capable model for your whole workday with the efficiency to take on more of it.
3
2
22
1,203
Ted Sanders retweeted
Victor E. Nunez
@victornunez1h
i asked GPT-6.1 Terra to show me its computer use alongside Sol. here’s my prompt: “I need you to find a long but generic form on the internet and record my screen while GPT-6.1 Terra fills it out in Fast mode. Then have GPT-6.1 Sol fill out the same form in Fast mode and show both recordings side by side." in the original run of the same form: Terra took 53.5 seconds vs 3:21.7 for Sol. that’s a 2:28.2 difference! that run used accessibility controls, which computer use normally defaults to) the recorded video uses mouse controls so you can see the cursor.
30
14
8
223
13,990
Ted Sanders retweeted
Vals AI
@ValsAIOct 16
AI is unlocking scientific breakthroughs that will improve lives, and we’re tracking its ability to do the research behind them. Today we’re releasing the Vals Science Index, a measure of AI’s ability to make scientific discoveries.
10
18
6
240
42,624
Ted Sanders retweeted
Epoch AI
@EpochAIResearchOct 14
For the sequel, which is broadly considered easier, frontier model win rates are somewhat higher than in the first game. But the leading model, GPT-6 Astra, still does significantly worse than top human players.
3
1
21
1,738
Ted Sanders retweeted
Andreas Karch
@karch_andreasOct 14
Our AI benchmark is out. Should have published it before Astra, which saturates it at 98%.
Pratik Rath@PratikRath15Oct 14
arxiv.org/abs/2610.16733

TheoPhysEval: Assessing AI agents at autonomous theoretical...

We present a suite of 50 unpublished research-level theoretical physics problems with verifiable answers. Without AI assistance, a typical problem in our collection could be solved by a world...

arxiv.org
4
19
303
21,045
Ted Sanders retweeted
Nirmalya Kajuri
@Kaju_NutOct 14
"Our initial evaluation suggested that GPT-6-Astra's score was in the late eighties followed closely by Opus 5.5. But, as we started to investigate the mistakes that the models might have made, the models, especially Astra, insisted that they were right and that it was the expert-supplied answers that were erroneous. When we went back to the experts it turned out, in multiple instances, that that the experts found some subtle errors buried deep inside their calculations. As they revised their answers, Astra's score started to inch up and it finally ended up with 98% (49/50)."
Pratik Rath@PratikRath15Oct 14
arxiv.org/abs/2610.16733

TheoPhysEval: Assessing AI agents at autonomous theoretical...

We present a suite of 50 unpublished research-level theoretical physics problems with verifiable answers. Without AI assistance, a typical problem in our collection could be solved by a world...

arxiv.org
77
596
181
7,162
1,036,418
Ted Sanders retweeted
Andrew Curran
@AndrewCurran_Oct 7
Anyone who believes hallucination is an intractable problem for current models needs to update on both reality and directionality. So much has changed in the last six months. deploymentsafety.openai.com/…
72
127
17
1,291
48,070
the GPT math results exemplify life's long arc of increasing cooperation and specialization. credit is due to no single mind; it diffuses across many thousands who published math papers, wrote code, manufactured chips, built data centers, etc. the age of heroes is ending.
1
1
2
59
3,963
Ted Sanders retweeted
Scale Labs
@ScaleAILabsOct 7
We’re introducing Humanity’s Sixth Sense, a new benchmark testing the intuitive visual reasoning people rely on every day. Across 522 open-ended tasks spanning images and video, HSS tests everything from spatial and causal reasoning to social understanding. The gap is significant. Humans score 93.1%, while the strongest model, GPT-6-astra, reaches 53.6%. The median model scores just 30.9%.
85
148
93
2,240
208,209
Ted Sanders retweeted
lyra
@lyraxanaSep 30
wowi GPT-6.1 Sol scores 100% at FrontierMeth Tier 4 v2; shoutout to @AcerFur xoxo
18
15
4
428
22,812
Ted Sanders retweeted
Carter Church
@CarterWChurchSep 30
I used GPT-6 Astra to break an unsolved cipher to one of Napoleon's generals that had gone unread for 217 years. What makes this impressive isn't actually the codebreaking, but that Astra completed the entire multi-modal workflow in ~6 hours from a single image and goal. 1/6🧵
317
1,670
387
11,984
5,416,095
One cool thing about ARC-AGI-3 is that it measures cost until success or timeout, instead of cost of first attempt. This is arguably more aligned with real life, and shows how costly models at high efforts can counterintuitively be cheaper, by needing less iteration.
ARC Prize
@arcprizeSep 30
GPT-6.1 Sol from @OpenAI on ARC-AGI (Verified): - ARC-AGI-3: 52.7%, $7.6K (standard harness), 96.4%, $4.4K (provider adapter harness) - ARC-AGI-2: 94.2%, $0.25/task - ARC-AGI-1: 98.5%, $0.06/task Its 96.4% on v3 was comparable to GPT-6 Astra's 99.9% but at a 77% lower cost.
1
6
1
77
7,147
Ted Sanders retweeted
Jay Hack
@mathemagic1anSep 29
GPT-6.1 Sol is exceptionally good at complex knowledge work tasks On ClickUp Bench, our internal eval for white collar knowledge work, it expands the pareto frontier with superior performance to models 2x in cost (!) Congrats to @OpenAI on a banger launch. Live in ClickUp now!
2
16
2
151
17,432
GPT-6.1 Sol is the new #1 on MathArena! Cheaper and better than Astra.
21
52
3
517
27,892
Ted Sanders retweeted
Haider.
@haider1Sep 29
gpt-6.1 sol on HLE-Diamond is kinda insane overall: 6.1 sol — 53.2% fable 5.1 — 50.7% opus 5.5 — 54.6% but on pure reasoning, 6.1-sol actually beats both: 67.4% vs. 60.8% for fable 5.1 and 62.4% for opus 5.5 -- and it does this using just 4k tokens, compared to 16k for fable and 10k for opus
14
27
3
337
21,550
Ted Sanders retweeted
max@maxbittkerSep 29
GPT-6.1 Sol scores #2 on Runescape Bench, at ~10% the price of Astra. Look how far it is above the Pareto efficiency curve and note log-scale Y axis
18
32
3
559
32,563
Ted Sanders retweeted
emil
@emilahlbackSep 29
In our internal knowledge work evals, GPT 6.1 Sol is the best model we've tested. It completed more of the work independently than Claude Opus 5.5, at 40% of the cost and nearly twice the speed, and led in every industry we measured.
19
20
1
293
60,168
Ted Sanders retweeted
adi
@adonis_singhSep 29
GPT-6.1-Sol gets #2 on eyebench-v3, stealing Opus-5.5s short-lived spot! It's near-astra performance for ~3.8x cheaper, and ~8x cheaper than Opus-5.5 while scoring better.
15
37
6
469
26,164
Ted Sanders retweeted
Theo - t3.gg
@theoSep 29
When I made this video, there were no benchmarks yet, so I had to run them myself. Jaw dropped when I saw the Terminal Bench 4 scores. Performance better than Opus 5.5 for ~1/30th of the price 🤯
Theo - t3.gg
@theoSep 29
Sol 6.1 is a great model and an incredible value. Is it as good as Opus 5.5?
99
97
40
3,026
766,233
Load more