erick

sup

[erick @ericc__ch]

@tan_stack tanstack gpu???

[erick @ericc__ch]

https://t.co/5wfnSNrxxi

Apparently I work whenever however I want lmao

Drifter: "The fallback when no stronger rhythm wins by enough: no clear night cluster, sunrise lean, lunch spike, or weekend tilt."

[erick @ericc__ch]

rate my p50 build

[erick @ericc__ch]

am i the only one reading wip as wee-ai-pee?

[erick @ericc__ch]

@kitlangton @EffectTS_ @opencode Why "streaming" in quotes?

[wukko @uwukko]

A fresh Brave install in 2026: sponsored ad wallpapers on new tab page by default (opt out). Brave VPN, News, Talk, Leo (AI), Rewards and other revenue-milking bloat is advertised/pinned by default. Analytics and "phoning home" by default. Google as default search engine in most regions by default. Sponsored search engines like Russian Yandex in CIS countries by default: https://t.co/nCK7ZErytG

Brave has an ad branch that handles advertising within the browser: https://t.co/EwJwKKgWxq. Brave does on-device ad targeting based on cohorts and interests, just like what Chrome used to do and what Google was largely hated for (remember FLoC?). This applies to additional (opt-in) rewarded ads, shipped as part of Brave.

Brave has injected referral IDs to crypto-related URLs entered into the omnibox in the past, intentionally, by design:
https://t.co/hovGMDz8Et
https://t.co/3wdWDEV85H
https://t.co/fo1mD75vAF

Brave also uses dark patterns to drive users away from turning off ads in their browser. For example, an article linked from the "opt out" button in the browser has a wall of text making excuses for ads before the actual steps needed to be taken to disable them: https://t.co/dka1EIMtbw

kind of hypocritical for brave to judge firefox for lesser bullshit, don't you think?

[erick @ericc__ch]

@threepointone @deepfates You play dark souls???

[erick @ericc__ch]

@antirez @mitsuhiko What the 🫠 thank god I've never (probably haven't) run into this

[erick @ericc__ch]

@ericzakariasson It's so good and it's so cheap I have a hard time depleting my composer usage on the $20 plan

Only wish I could use it outside cursor

[erick @ericc__ch]

@thdxr @adamwathan dax you liar you said you were dumb

[erick @ericc__ch]

@Sauers_ Damn the speedrun needs multiplayer??

[erick @ericc__ch]

might be a hot take but gemini 3.5 flash is actually usable for doing actual "agentic coding" work now

tho I wouldnt use it if I have to pay for it
I guess its not a hot take after all

[Artificial Analysis @ArtificialAnlys]

Cursor's new Composer 2.5 takes third on the Artificial Analysis Coding Agent Index and is ~10-60x lower cost than the higher-effort Opus 4.7 and GPT-5.5 variants above it. This release puts Composer among the leading coding agent models, something that wasn’t clear for past releases

@cursor_ai has released Composer 2.5, the latest model in its Composer line. Composer 2.5 scored 62 on our Coding Agent Index, a 14 point gain over Composer 2 (48). This puts it in third place of our tested agents, behind only Claude Opus 4.7 (max) in Claude Code (66) and GPT-5.5 (xhigh reasoning) in Codex (65). These cost $4.10 and $4.82 per task respectively, ~10x the cost of Composer 2.5 Fast ($0.44) and ~60x the cost of Composer 2.5 standard ($0.07).

Key results for Composer 2.5 in Cursor CLI:

➤ Cost-quality Pareto frontier: At $0.07 (standard) and $0.44 (Fast) per task, Composer 2.5 is cheaper than every other agent scoring above 60 on the Index. Medium-effort peers cost $1.24–$2.21 per task; higher-effort variants land 3-4 points above at $4.10–$4.82

➤ Per-benchmark gains vs Composer 2: +35 points on SWE-Bench-Pro-Hard-AA (12% → 47%), +2 points on Terminal-Bench v2 (64% → 66%), and +3 points on SWE-Atlas-QnA (69% → 72%). At 47%, Composer 2.5's score on SWE-Bench-Pro-Hard-AA is comparable to Claude Opus 4.7 (max) in Claude Code

➤ Among the fastest coding agents: Composer 2.5 Fast runs at an average wall time of 6.7 minutes per task, the third-fastest agent on the Artificial Analysis Coding Agent Index, behind only Claude Opus 4.7 (medium) in Claude Code (5.8m) and GPT-5.5 (medium) in Cursor CLI (6.2m)

➤ Fast mode enables better responsiveness at 6x pricing: Fast runs 30% faster than standard Composer 2.5, but is ~6x the cost per task ($0.44 vs $0.07). Token pricing is 6x higher for Fast: $3.00/$15.00 vs $0.50/$2.50 per million input/output tokens

Model details:

➤ Base model: Continued training on @Kimi_Moonshot's open weights Kimi K2.5 as with Composer 2, with Cursor reporting ~85% of total compute from its own additional training and reinforcement learning

➤ Pricing: $0.50/$2.50 per million input/output tokens for the standard variant; $3.00/$15.00 for the Fast variant (the default in Cursor)

➤ Available exclusively in Cursor: both Cursor IDE and Cursor CLI, an externally accessible API is not available

Congratulations @cursor_ai and @mntruell on the impressive release!

[erick @ericc__ch]

@thdxr Cool now I know about heyapi 😋

[erick @ericc__ch]

@badlogicgames YES

[erick @ericc__ch]

@nahcrof I'd subscribe right away if I can use it practically unlimited (in my case more than 2k req/day) for the cheapest sub

Might be too much to ask lol sorry

[erick @ericc__ch]

@Nx4p0n +1 at least. Cool art!

[erick @ericc__ch]

@MichaelArnaldi @devagrawal09 Maybe I should learn to do it too