Felix Rieseberg ✓
@felixrieseberg · San Francisco · joined 03 Apr 2009
I build things @AnthropicAI, Co-Maintainer https://t.co/g4potti8nq
72 025Followers
721Following
2 853Posts total
1.4MViews on collected posts
Latest posts
Looking for new posts…
@felixrieseberg So Grok High is better than the new fable unless you exclusively run xhigh or max on it and pay more.
1.7K views · 78 likes
· 0 reposts · 0 replies
Open on X →
Claude Fable 5.1 is available today in Claude, Claude Code, Cowork, and via the API, plus Bedrock, Vertex AI, and Azure.
Mythos 5.1, which is the same model with more permissive safeguards for certain tasks in high-risk, dual-use domains such as biology and cybersecurity, goes
19.2K views · 94 likes
· 3 reposts · 0 replies
Open on X →
Let's talk more about writing. Where earlier models overused bullets and bold in chat, Fable 5.1 uses bold less, and is less likely to reach for headers, lists, or quotation marks.
There's more to be done (and we will) but Fable 5.1 also respects my style instructions better,
32.6K views · 138 likes
· 8 reposts · 13 replies
Open on X →
Science got a big jump. Everyone's been talking about math for the last weeks and I think other disciplines will soon see similar breakthroughs.
On Terminal-Bench-Science (@Stanford-led, tasks written by working scientists), Fable 5.1 scores 52.6%, more than double Fable 5 https
36.6K views · 178 likes
· 9 reposts · 13 replies
Open on X →
Today, we're releasing Fable 5.1 and Mythos 5.1. While the model is getting slightly better at a lot of things, I expect many users to appreciate that Fable 5.1 solves agentic coding tasks at roughly half the cost of Fable 5.
I personally also really like it's writing style, htt
709.8K views · 3.5K likes
· 140 reposts · 213 replies
Open on X →
Happy Opus 5 launch day! It's a good model - on multiple benchmarks, it's SOTA and beats Fable 5, especially long-horizon agentic tasks. On the rest, it gets pretty darn close.
I bet that it'll become many people's favorite workhorse for serious work in Claude Code, Cowork, and
14.3K views · 197 likes
· 6 reposts · 13 replies
Open on X →
New feature in Language Model Builder: You can now import base models.
The results are a bit of an overfitting fever dream, but I've also added iMessage group chats as a fine-tuning data source.
Come, train a model based on your silly group chats: https://t.co/nLjFfpczbV https
13.1K views · 122 likes
· 5 reposts · 9 replies
Open on X →
@felixrieseberg This is soo cool! You should take more time for these things!!
2.9K views · 14 likes
· 0 reposts · 0 replies
Open on X →
It all takes less time than you might think, too. I kept he introduction as tight as possible. It probably takes about 90 minutes to read.
In about a day, you can do pre- and post-training on a model that'll write coherent responses.
26.4K views · 132 likes
· 0 reposts · 4 replies
Open on X →
Have you built a language model? You should. It's so much fun to chat with something you made.
Anyone can do it, too. I made an app that teaches the fundamentals and gives you everything you need to build your own: https://t.co/nLjFfpczbV https://t.co/akIMnTzOEG
569.6K views · 3.1K likes
· 273 reposts · 97 replies
Open on X →
Something I’m excited about: Free premium Claude and Cowork for K-12 educators and teachers in the US, together with a library of teaching skills.
To find out more, check out https://t.co/O20Lnyd3EK
14K views · 109 likes
· 4 reposts · 12 replies
Open on X →
Against accounts of the same size
6 posts from the last 90 days, next to the 10K–100K follower range. shown widely, but few of those viewers react.
Median views25 894this account924median for 10K–100K
Reach, %35.95%this account3.62%median for 10K–100K
Engagement, %0.55%this account1.52%median for 10K–100K
| Metric | This account | Median for 10K–100K | Ratio |
|---|---|---|---|
| Median views per post | 25 894 | 924 | 28.0× |
| Reach (views ÷ followers) | 35.95% | 3.62% | 9.93× |
| Engagement rate | 0.55% | 1.52% | 0.36× |
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
14K14 Jul
569.6K21 Jul
26.4K
2.9K22 Jul
13.1K
14.3K24 Jul
709.8K1 Sep
36.6K
32.6K
19.2K
1.7K
Last 11 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.
Engagement rate per post
0.89%14 Jul
0.61%21 Jul
0.52%
0.48%22 Jul
1.05%
1.51%24 Jul
0.56%1 Sep
0.55%
0.50%
0.51%
4.61%
Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.
What the audience does
Likes46.4%7 602 in total
Reposts2.7%448 in total
Replies2.3%374 in total
Quotes1.0%172 in total
Bookmarks47.5%7 791 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.
Similar accounts
瞎玩菌 ✓
@Blind___Gamer
144.1K followers
· 36.5K posts
Matt Ridley ✓
@mattwridley
144.1K followers
· 47.4K posts
Diego Muñoz
@diegomunoz75
144.1K followers
· 47.1K posts
PLOS
@PLOS
144K followers
· 25.1K posts
NPC Moments‼️ ✓
@npc_moments
144K followers
· 2.1K posts
Geezy..
@MsMyaG
144K followers
· 345.4K posts
Snayzy ✓
@Snayzyy
144K followers
· 4.6K posts
𝙋𝙍𝙔𝙉𝘾𝙀. ✓
@PrinceduParc
144K followers
· 64.8K posts