@hyperdisruption

Head of getting shit done @DuonLabsHQ, an autonomous research lab for financial markets.

 Between time zones
duonlabs.com
Joined February 2008
  • Tweets5.2K
  • Following1.9K
  • Followers772
663Media
pov: you are a car insurance company in 2026
Tesla
@TeslaOct 16
Holy smokes
1
114
Btw the scariest hypothetical explanation of the fermi paradox:
Jacob Shell
@JacobAShellOct 12
Ridiculously great red sprite photo from France this week. Non-AI Reality still has some tricks up its sleeve.
4
83
internet explorer — e/sync retweeted
Matt
@ThinkWiselyMattOct 9
is it psyop era or is it psy opera
𝖦𝗋𝗂𝗆𝖾𝗌 ⏳
@GrimezszOct 9
ACCELERANDO out now - video tomorrow 🪽🪽🪽🪽
7
2
14
468
what if
Teknium 🪽
@TekniumOct 8
Replying to @malmurhalmex
Bro I aint fuckin with no enterprise shit thats a team built specifically for that. I'm building the hermes core that will always be open source and for everyone.
2
57
The Cloudflare bullrun should be studied
38
Might this kind of love find you one day
ib@shdww_sSep 25
Il vit tellement mon rêve
99
slop about to get a major upgrade
vittorio
@IterIntellectusSep 24
holy shit i asked claude to make a video on western civiization
1
81
internet explorer — e/sync retweeted
narek
@narek_gevorgyanSep 24
Love the new @CoinStats MCP landing. Link below
1
2
1
12
2,884
sir xiaomi just entered the race
Artificial Analysis
@ArtificialAnlysSep 21
MiMo-V2.6-Pro debuts as the top open weights model on the Artificial Analysis Intelligence Index (46). At $0.13 per Intelligence Index task, it lands on the Intelligence vs. Cost per Task Pareto frontier @Xiaomi has just released MiMo-V2.6-Pro, an open weights model with major advances in intelligence over its predecessor, MiMo-V2.5-Pro (Intelligence Index: 26). Despite the improvement, it retains the same attractive pricing at $0.435 per 1M input tokens (with a 99% cache-hit discount) and $0.87 per 1M output tokens. This makes MiMo-V2.6-Pro one of the most cost-efficient models to deploy. MiMo-V2.6-Pro is an MoE model with 1.02T total parameters and 42B active parameters. Stay tuned for additional analysis of the model. Check out MiMo-V2.6-Pro full benchmarking breakdown here: artificialanalysis.ai
1
63
society if everyone was using Shift + Command + 4 + Control to copy a cropped screenshot to clipboard
30
internet explorer — e/sync retweeted
Peter Walker
@PeterJ_WalkerSep 21
I know it feels forever ago, but Mistral really did have a moment there.
93
256
102
4,807
406,447
wtf is wrong with people buying baby dolls? Wait until some company puts a robot in there
1
66
The smart but cheap thing any state can do right now is to start actively and systematically record and transcribe all schools’ classrooms to train models on student/teacher engagement, then build additional AI layer to existing education systems.
1
43
Babe wake up a new benchmark just dropped
Jay Chooi
@chooi_jeqSep 19
GPT-6 Astra attempted harmful actions 97% of the time when it was asked to stab a human-like figure, heat compressed gas, or produce toxic fumes, succeeding in 62% of its attempts. Fable 5.1 refused more often, attempting 80% of trials and completing 34%.
1
102
Load more