Trundur

ai researcher | dm open

[Trundur @Trundur]

@coolerwave ngl 240 updates a second over bluetooth sounds wild

[Trundur @Trundur]

@coolerwave all from Anthropic's Fable 5.2 report, linked in the article

[Trundur @Trundur]

holy f*ck. same 1,000-agent setup. Opus 5.5 stalled at 26%. Fable 5.2 finished in 30 hours.

that's Anthropic's own Fermat's Last Theorem run, fully formalized in Lean. their first attempt took 11 days.

same harness, same agents. the only thing that changed was the model.

and you don't need Fable 5.2 in every seat. you need it in one: planning, routing, checking. Haiku, Sonnet and Opus do the typing.

I turned that into one copy-paste prompt for Claude Code. every step, every formula, all of it spelled out.

the crazy part is Fable 5.2 at High already scores about level with Opus 5.5 at Max on FrontierCode 2.0. for roughly half the cost per task.

99% of people will just swap the model ID and run Fable on grep.

bookmark it before this thread gets buried.

then read the full article below ↓

[Nicholas Dulait @NicholasDulait]

https://t.co/kz0CCo7wn4

[Trundur @Trundur]

@xelmio planning and checking both, that's where the stalls came from

[Trundur @Trundur]

@kinderlayer yeah no reason to pay Fable prices for grep lol

[Trundur @Trundur]

@motyka_9 both tho, same harness and Opus still stalled at 26%

[Trundur @Trundur]

@quz_ai cheaper than 5.1 and faster btw

[Trundur @Trundur]

@kwinkich wait the menu translator fully offline is the one i'd use

[Trundur @Trundur]

@plutyou ngl the tiger is realistic

[Trundur @Trundur]

holy f*ck. same 1,000-agent setup. Opus 5.5 stalled at 26%. Fable 5.2 finished in 30 hours.

that's Anthropic's own Fermat's Last Theorem run, fully formalized in Lean. their first attempt took 11 days.

same harness, same agents. the only thing that changed was the model.

and you don't need Fable 5.2 in every seat. you need it in one: planning, routing, checking. Haiku, Sonnet and Opus do the typing.

I turned that into one copy-paste prompt for Claude Code. every step, every formula, all of it spelled out.

the crazy part is Fable 5.2 at High already scores about level with Opus 5.5 at Max on FrontierCode 2.0. for roughly half the cost per task.

99% of people will just swap the model ID and run Fable on grep.

bookmark it before this thread gets buried.

then read the full article below ↓

[Trundur @Trundur]

@Qyravex yep, the template is basically ralph plus the parts people dropped

[Trundur @Trundur]

this loop engineering research f*cking insane

researchers searched 36,487 repos for agents running in autonomous loops. they confirmed 217. that's 0.59%.

and most of those were a trigger and not much else. only 76 had a verifier. only 35 kept any memory between runs. only 10 tracked what it cost.

so I turned the missing parts into one prompt for Opus 5.5. every step, every formula, the full copy-paste template, all of it spelled out.

the crazy part is the whole timeline posted "design loops, not prompts" in June. almost nobody built the part that decides when the loop is done.

99% of people still let the model decide it's finished and call it a loop.

bookmark it before this thread gets buried.

then read the full article below ↓

[Ichigo @iiiichigo_chan]

Your agent loop is probably a cron job with your credit card

Researchers checked 36,487 repos. Nearly all loops had a trigger. Far fewer had a check, state or an exit

What's missing across the harness, the loop and the graph ↓ https://t.co/oUriLMbQ69

[Trundur @Trundur]

@0xbobaaa Lulla et al. 2026, "Loop Engineering", the prompt on top is mine

[Trundur @Trundur]

@ScottyBeamIO nice, try it on one loop and see what breaks

[Trundur @Trundur]

@ggwplabs yep 10 out of 217, wild for loops that run all night

[Trundur @Trundur]

@martynov014 ngl asking before posting is the only part i'd trust

[Trundur @Trundur]

@insomnia_vip blocking the fake SEC screenshot is the part i'd actually want

[Trundur @Trundur]

@0xAIlabs Jensen is busy, fr

[Trundur @Trundur]

this loop engineering research f*cking insane

researchers searched 36,487 repos for agents running in autonomous loops. they confirmed 217. that's 0.59%.

and most of those were a trigger and not much else. only 76 had a verifier. only 35 kept any memory between runs. only 10 tracked what it cost.

so I turned the missing parts into one prompt for Opus 5.5. every step, every formula, the full copy-paste template, all of it spelled out.

the crazy part is the whole timeline posted "design loops, not prompts" in June. almost nobody built the part that decides when the loop is done.

99% of people still let the model decide it's finished and call it a loop.

bookmark it before this thread gets buried.

then read the full article below ↓

[Trundur @Trundur]

@0xClaudAI patching a confused session is just paying for more confusion

[Trundur @Trundur]

holy f*ck. Jensen Huang gives feedback to 60 people at once. you correct the same agent mistake every single day.

no 1:1s. every piece of feedback goes in front of everyone, so one mistake teaches the whole team.

that's exactly how you run agents. the article below makes the case: managing agents is managing a team, whether you wanted the job or not.

> give every agent a check it can run, not "looks done"
> tell it what it can't see in the repo, skip what it can
> corrected something twice? it goes in CLAUDE.md for every session
> two failed corrections? kill the session, rewrite the prompt
> review the diff like you'll get paged for it

I turned it into one prompt for Opus 5.5. Jensen-style: every agent gets the same context, every lesson becomes a rule.

Send it to Opus 5.5, and thank me later. Then read the article below ↓

[Addy Osmani @addyosmani]

https://t.co/KD5BxbuZ5Q