Findr

trying every new ai tool so you don't have to. some are even good.

[oracle @quietoraclex]

holy sh*t, split that five-department Opus studio into an Agent Team and ONE Opus session quietly becomes 5 Opus bills...

with teams on, any subagent Claude names becomes a teammate, asked or not. and every teammate runs on the lead's model unless you say otherwise

1 Opus lead + 4 teammates, each reading 1M tokens and writing 100K:
→ all Opus 5.5: ~$30
→ Sonnet 5.5 teammates: ~$18
→ Haiku 5.5 teammates: ~$6.60
(list prices, no cache)

one env var fixes the default:
CLAUDE_CODE_SUBAGENT_MODEL=sonnet claude --model opus

→ Opus leads and reviews
→ any teammate without a model of its own runs on Sonnet
→ name haiku in the spawn prompt for read-only grunt work
→ teammates copy the lead's effort too, so check /effort status

you picked Opus once. the team picked it 4 more times.

one Claude Code bill trap every night. tomorrow's is already in the queue.

[Findr @aifindr]

@AIDuckLabs yes

[Findr @aifindr]

DEEPSEEK BURNED 11X MORE TOKENS THAN MINIMAX ON THE SAME PROMPT.
IT STILL PAID 54 CENTS. THE CATCH IS THE CACHE.

round 1. same prompt, same task, the two new flash models:
> deepseek v4.1 flash: 1h 28m, 89.5M tokens, $0.54
> minimax m3.1 flash: 1h 06m, 8.1M tokens, $0.22

minimax won the round. 22 min faster, 2.4x cheaper.
but look at that deepseek token count.

at the normal input price, 89.5M tokens is $26.85.
deepseek charges $0.006 per 1M on a cache hit.
89.5M × $0.006 = $0.54. the exact bill.

so the agent spent almost the whole run rereading its own context from cache.

the price list isn't your bill. your cache hit rate is:
> keep the system prompt and file order stable
> don't let the agent rewrite early context mid run
> every cache miss bills at full price

today deepseek open sourced deepinfer, the engine it says runs v4.1 flash at scale. the whale's kitchen is public now.

the cheapest model is the one that remembers what it already read.

one prompt, two models, real receipts. a new fight every day.

[Hamel Husain @HamelHusain]

New blog post 🎉: How To Use Decision Models For Evals

In this post, we show you how to choose a decision model for your task, tune its threshold (don't assume the 0.5 default is good!), and interpret its scores.

Lots of cool interactive charts here. https://t.co/uYAfte9Kb5

[Findr @aifindr]

@quietoraclex maybe tomorrow

[Findr @aifindr]

@insomnia_vip same. whichever drops first goes straight into the ring

[Findr @aifindr]

@diamai_ 100%. the test only posted one total, so i backed the cache share out of the bill. a 3 line receipt would settle it for good

[Findr @aifindr]

@Kallias_X headline price is the menu. cache hit rate is the receipt

[Findr @aifindr]

@0x_xena on this test, yes. 22 min faster and 2.4x cheaper. round 2 might go differently

[Findr @aifindr]

@itsarcvale opus writing, haiku fact checking it is the most realistic office setup i've seen all week

[Findr @aifindr]

@tomniy_bes @magicmarkets same lesson as ai pricing. the number on the screen is never the bill

[Findr @aifindr]

@0x_xena 200 agents is easy. 200 agents that don't all reread the same context at full price is the real flex

[Findr @aifindr]

@Principia_X the security block is the only part people skip. https://t.co/Cf0V4dutWo and https://t.co/cxBQ8L6H4R before you touch anything on this list

[Findr @aifindr]

@alxqwrk warranty departments are about to have the worst year

[Findr @aifindr]

@0xVeyyro night shift is exactly where the slow cheap model belongs. nobody's awake to care it took 3x longer

[Findr @aifindr]

@Synthexfr everyone wants the second brain. nobody wants the seatbelt

[Findr @aifindr]

@0xOrionVega very good stack bro

[Findr @aifindr]

@velesxbt so the model's main skill is losing to the buy button. love that someone actually tested it

[Findr @aifindr]

@rugikkk the best line here is "the only button that moves money is mine". six agents, zero spending rights. that's how every agent desk should be built

[Findr @aifindr]

@quietoraclex really useful, thanks

[Findr @aifindr]

@itsarcvale ahahahhhaahhah

[Findr @aifindr]

MINIMAX JUST SHIPPED A MODEL 26X CHEAPER THAN OPUS 5.5.
AN EARLY BUILD WAS ALREADY #1 ON OPENROUTER UNDER A FAKE NAME.

for 10 days the #1 model on openrouter was a stealth model called space bunny.
56.7T tokens. nobody knew whose it was.

today it got a name: minimax m3.1 flash.

per 1M tokens, in / out:
> opus 5.5: $4 / $20
> deepseek flash: $0.30 / $1.20
> minimax m3.1 flash: $0.15 / $0.60, flat, no peak hours

the whale's weekend price is now someone's every day price.

what $0.60 buys, by minimax's own table:
> automation bench: 62.2. opus 5: 50.3
> DeepSWE: 73.7. opus 5: 74.0
> one tester's one shot 3d pagoda: 50K tokens, about 3 cents

the catch, same table:
> terminal bench: 38.1. opus 5: 51.8. claude code lives in the terminal
> the whale still wins DeepSWE at 77.0
> weights land oct 23. free on openrouter for 72 hours, then it bills

opus didn't get worse this week.
it just got expensive.

[MiniMax (official) @MiniMax_AI]

MiniMax M3.1 Flash: Space Performance + Bunny Speed

🚀🐰 Yes, Space Bunny - which has topped global leaderboards with 200T+ tokens - was an earlier version of M3.1 Flash.

💰Redefining the performance economics of agentic AI. M3.1 Flash achieves frontier-level intelligence across key tasks such as Cowork (Automation-Bench: 62%; ALE: 30%), Coding (DeepSWE: 74%) and Visual Reasoning (BabyVision: 91%), at 1/3 the cost of M3 and 3x the speed.

API pricing (flat rate): $0.15 / $0.60 per 1M input/output tokens, with cache reads at $0.015 / 1M tokens.

Model weights will arrive on Oct 23.

API: https://t.co/fHRdSV73Hr
M Plan: https://t.co/XfJWSyauda
MiniMax Code for desktop: https://t.co/SGmLTI2HQm