"͏ ͏" (@DownTimeMachine) header
"͏ ͏" (@DownTimeMachine) profile picture

"͏ ͏"

@DownTimeMachine

Dada scientist. Generative art, pen plotters, LLM swarms.

xn--ho8hhgk5byo.y.at/
Joined April 2011
  • Tweets5.6K
  • Following6.1K
  • Followers1.5K
Image from "͏ ͏" (@DownTimeMachine)
Da7em (@Da7_Tech) profile pictureDa7em
@Da7_Tech21h
I think Hermes Agent offers way more value than Grok Bot.
1
2
42
Image from "͏ ͏" (@DownTimeMachine)
1
1
34
It also is super easy to switch models on the fly
Image from "͏ ͏" (@DownTimeMachine)
1
10
I mostly use Kanban to keep my agents working. My app is synchronized to the main Kanban board that all agents work from, so when they're working I can sort in app by status (or specific agent or server or project)
Image from "͏ ͏" (@DownTimeMachine)
1
17
But it's actually X
Image from "͏ ͏" (@DownTimeMachine)
48
Replying to @gakonst
They're looking for daylight savings time gurus at a crypto company!
Image from "͏ ͏" (@DownTimeMachine)
34
Replying to @SHL0MS
@SHL0MS
Image from "͏ ͏" (@DownTimeMachine)
2
34
Image from "͏ ͏" (@DownTimeMachine)
1
43
Image from "͏ ͏" (@DownTimeMachine)
1
1
Step 5 Preview is free and nearly matches paid models on routine agent tasks: 46/48 vs 48/48 for gpt-6.1-sol and sonnet-5.5. On harder replays of real work it drops off: avg score 4.5/10 vs ~6.8. Nous's Hermes Index (33.89, tied with GPT-6 Luna) agrees.
Image from "͏ ͏" (@DownTimeMachine)
Nous Research (@NousResearch) profile pictureNous Research
@NousResearchOct 9
Step 5 Preview from @StepFun_ai is free on Nous Portal for the next week 600B total / 27B active MoE, 1M context, with vision Step 5 Preview scored 33.89 on Hermes Index, the same score as GPT-6 Luna
1
96
Load more