François Chollet ✓

@fchollet

Co-founder @ndea. Co-founder @arcprize. Creator of Keras and ARC-AGI. Author of 'Deep Learning with Python'.

723 330Followers

827Following

25 659Posts total

684.7KViews on collected posts

Against accounts of the same size

6 posts from the last 90 days, next to the 100K–1M follower range. shown to more people than peers of the same size.

Median views29 352this account6 139median for 100K–1M

Reach, %4.06%this account2.61%median for 100K–1M

Engagement, %1.13%this account1.39%median for 100K–1M

Metric	This account	Median for 100K–1M	Ratio
Median views per post	29 352	6 139	4.78×
Reach (views ÷ followers)	4.06%	2.61%	1.55×
Engagement rate	1.13%	1.39%	0.81×

Others in this range → Compare with another account → How these benchmarks are built →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

579.1K3 Sep

44.2K

29.2K

29.5K

471

2.2K4 Sep

Last 6 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

1.16%3 Sep

1.10%

0.98%

1.67%

0.21%

7.55%4 Sep

Reactions — likes, reposts, replies and quotes — divided by views. Median for 100K–1M accounts is 1.39%.

What the audience does

Likes68.4%6 929 in total

Reposts8.1%823 in total

Replies1.8%182 in total

Quotes2.3%229 in total

Bookmarks19.4%1 961 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.

Latest posts

@fchollet Ineed it is, as I predicted, but there is a caveat, what is the compute/power rational to achieve it? how about #NoGPU #NoLLM to solve ARCAGI3 this is what #Neuraxon is a real bioinspired brain architecture https://t.co/NQ4JiO9dK1 2.2K views · 126 likes · 36 reposts · 0 replies 04 Sep 2026 @fchollet how'd you determine these OOM differences? 471 views · 1 likes · 0 reposts · 0 replies 03 Sep 2026 When we released ARC 3, I got asked, "when do you think a frontier model will saturate it?", and I answered "in about a year, though it depends on how much it gets explicitly targeted" That was 6 months ago, so the progress that Astra represents happened about 2x faster than I 29.5K views · 437 likes · 37 reposts · 10 replies 03 Sep 2026 Benchmarking AI systems is a continual process that co-evolves with the models. New benchmarks challenge AI capabilities with emerging questions to shape the directions and feedback signal of the research process. Then they adapt as models progress, targeting the residual between 29.2K views · 263 likes · 10 reposts · 13 replies 03 Sep 2026 Many of you will ask, "if it saturates ARC 3, is it AGI?" We're not making this claim. All we know about the system so far are its benchmark scores. When we launched ARC 3, and in every presentation we made about it, we were very insistent on one thing: solving it is not proof 44.2K views · 432 likes · 26 reposts · 15 replies 03 Sep 2026 GPT-6 Astra represents a step-function change in model capability for interactive reasoning problems. It scores 66% on ARC-AGI-3 using our standard harness, and nearly 100% with a continuous conversation harness and custom compaction, at a cost of roughly $360 per game. In fact, 579.1K views · 5.7K likes · 714 reposts · 144 replies 03 Sep 2026