Epoch AI

@EpochAIResearch77k followers

Investigating the trajectory of AI for the benefit of society.

Open on X

Top outlier

184×

Best organic post

Peak views

Single post

Bookmarks

4.6k

Across 8 organic posts

Last scan

Oct 16

Community cache

Posts 8 posts

#	Date	Post	↓ Outlier	Views	Likes	Reposts	Bookmarks
1	Sep 22	AI is getting cheaper more quickly than any other transformative tech in history. At a given level of performance, cost has fallen ~47%/quarter since 2023. That’s 4× faster than DN	184×	2M	8.8k	1.4k	4.1k
2	Oct 15	Gemini 4 Argon scores 165 on the Epoch Capabilities Index (ECI), on par with Fable 5.1 (165) but behind Opus 5.5 (167) and GPT-6 Astra (166). While behind the ECI frontier, Argon i	4.9×	53k	604	60	74
3	Oct 8	Can AI automate Epoch? We're introducing Epoch Automation Reports to evaluate frontier models on realistic, open-ended tasks drawn from our own work. Claude Fable 5.1 and GPT-6 Ast	2.4×	26k	383	34	102
4	Oct 15	The New York Times visualizes the staggering amount of math needed to train AI models, using data from Epoch's research teams 📊 Our training compute estimates: The Epoch Capabilit	2.1×	22k	127	13	48
5	Oct 9	In 3 of the 18 math subfields we track, more than half of arXiv papers by established authors now acknowledge using AI. Rates have increased rapidly: in differential geometry, the	2.0×	21k	323	34	80
6	Oct 8	A new architecture for GPT-6.1 Sol? OpenAI halved its cached-input price compared with GPT-6 Sol. Our measurements show it also handles long prompts faster.	1.5×	17k	356	27	56
7	Oct 14	Six months ago, frontier AI models could barely beat Slay the Spire. Now, according to our new SpireClimb benchmark, they win 90%+ of runs at the easiest difficulty, and up to 20%	1.4×	15k	349	23	56
8	Oct 14	While frontier models are getting better quickly, their win rates on the hardest difficulty in both games are still far below those of top human players, with GPT-6.1 Sol leading i	0.4×	4.4k	30	1	3
9	Oct 15	Argon’s performance varies across domains. It tops some knowledge, agentic, and image benchmarks, while lagging behind in math and software engineering, ranking 8th on Math ECI and	0.4×	3.9k	48	2	2
10	Oct 15	Learn more about the Epoch Capabilities Index (ECI) and the benchmarks that feed into it:	0.3×	3.5k	23	1	1
11	Oct 9	Explore how AI acknowledgment rates vary by math subfield, use case, and provider:	0.3×	3.4k	16	0	4
12	Oct 8	This isn’t conclusive evidence of a new architecture, but it suggests something has changed in how GPT-6.1 Sol handles long contexts. This is a follow-up to our earlier report on l	0.3×	3.3k	29	2	3
13	Oct 16	How far behind the closed frontier are open-weight cyber capabilities? About 6–7 months. Zai’s GLM-5.3 is the highest-scoring open-weight model on our Cyber ECI. Its score of 160 b	0.3×	2.8k	74	10	10
14	Oct 8	For example, GPT-6 Astra ran an experiment exploring why AI agents fail to learn with practice. It set AI agents’ token budgets too low. Instead of treating this as a mistake, it r	0.2×	2k	24	0	2
15	Oct 8	Over time, we’ll expand this task suite, retire saturated tasks, and publish updated findings as new models are released. Read @kellyhongsn’s full report:	0.2×	2k	26	0	6
16	Oct 14	For more details and results, check out the full article.	0.2×	1.9k	14	0	2
17	Oct 14	The visual interface could be limiting performance, but frontier models don’t seem severely hampered by it, given they win 90%+ of the time on the easiest difficulty. We think meas	0.2×	1.9k	10	0	1
18	Oct 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 significantl	0.1×	1.6k	19	1	0
19	Oct 14	In SpireClimb, models play the game just like humans, through a graphical interface. Just a few months ago, we observed models in a similar setup fail to read relevant on-screen te	0.1×	723	11	0	0
20	Oct 14	Now, the best models carefully gather strategic information and execute smart plays smoothly. We’re planning to soon upload over a thousand videos that show the model gameplay we r	0.1×	695	13	0	0
21	Oct 16	Read the full Data Insight and methodology:	0.1×	692	5	0	1

Scan another creator

Run this on any account.

XOutlier builds this table for every public profile you open on x.com, with live data and every post, not just the top 50.

Add to Chrome Download .zip