BlockedPath

@screenslaver @zekramu @verymidengineer Right click

@zekramu @BlockedPaths @verymidengineer Maybe Im stupid but herdr didn’t seem to let me rename the tabs to a friendly description, making navigation no better (worse really)

in 6 months all of tech twitter will be talking about pi/omp supremacy

Fable 5 is dead. We just resurrected it — cheaper, open and you hold the keys. OpenRouter dropped Fusion 48h ago and broke the internet. We tested it hard. The synthesizer is insane for deep research… but absolute dogshit for coding. So we fixed it. Meet OrcaRouter.ai DSL — the version you actually own. One prompt → fans out to any panel you want → judge + synthesizer → one god-tier answer. But unlike black-box slugs, you control the entire graph in YAML. Fable 5 level intelligence… without waiting for Anthropic to turn it back on 🧵👇

M3's free on @0G_labs 0G compute for three days, starting today. Amazing chance to throw a long-running task at it 👇

Best fight I’ve seen in a while, Justin dog walked him. That knee to the ribs at the end was brutal.

@eersnington Same here, it feels so nice!

herdr is amazing. there hasn't been a lot of tools like it where i went "holy shit" in the first few mins

@0xSero It’s alright, I would say omp is better.

I can't believe I've been grokked. I have been using Grok build with my local models and I can't help but say the harness is phenomenal. So slick and smooth, so fast.

@loktar00 @usr_bin_roygbiv About time!

I'm ashamed to say (no not that I use Windows😂) that I'm just now finally installing omp after reading / liking / reposting @usr_bin_roygbiv for so long regarding it. Finally got time to breath for a minute today and install it!

@Da7_Tech And im almost positive this is just a fork of opencode

@Da7_Tech I have this problem too, some serious token leakage happening.

What a disaster. This genuinely goes against everything people expect from the values behind open-source models. Yesterday, Zcode was released — ZAI’s own app, built to work with their GLM models. I tried it on the Pro plan, the same Pro plan they describe as “15x Claude Code.” One single task consumed 60 million tokens from my limits in less than an hour, burning through the entire 5-hour limit without even completing the task. And the worst part? Their own app clearly showed the task used under 2 million tokens. But according to my tests, ZAI still deducted around 60 million tokens from both the 5-hour limit and the weekly limit. That strongly suggests cached tokens are being counted against me as real usage. So in practice, I am not even getting 5% of the value of my subscription. The same task was completed by Codex and MiniMax on their $20 plans. That says a lot about ZAI’s current infrastructure. I really hope they fix this, because I do want their models to succeed. I believed in the project enough to subscribe for a full year. But so far, for more than a month, I have not been able to actually benefit from my subscription. Let’s see what happens in the next few days.

@herdrdev Dropped cmux for this, issssa nice.

herdr 0.7.0 is out, and it's a major one: it introduces plugins! the idea is simple: herdr stays lean, and everything custom gets extended through plugins. shareable, scoped, built however you want, to fit your own flow. with this release we're also shipping a few examples of what the plugin system can do. first up: a telegram plugin. herdr already controls your agents and knows their status, so the plugin just hooks into agent events and pings telegram the moment one needs you. notification lands → `herdr --remote` or ssh from your phone → straight back to the agent that needs you.

@oleksoleksoleks @usr_bin_roygbiv Opencode GO, limits

@BlockedPaths @usr_bin_roygbiv Wait it's free? What are the limits

I had no freaking clue cloudflare provided models, just got some more compute with kimi 2.7 coding. Another win for the old .edu email

@notKartikk You’re going to hate me but more student freebies 😂😂😂

@BlockedPaths 😂😂

aww man why is this so true

@forloopcodes And the token usage on a new session is so lean and clean. Pi is goated

pi is crazy fast and you would not believe me until you try it yourself, i've stopped praising opencode, claude code and codex after a single day of using pi. it just works, as intended. no agent harness can even touch this beast. used it for like 3 hours and now i am addicted.

@mr_r0b0t My same thoughts, and I can’t absorb anything. Hitl is still really important, especially when shit hits the fan lol.

@BlockedPaths I actually prefer the slower as well! I don't think we're at the point where the human can be completely removed from the loop. Imagining this burning tokens trying to achieve some workaround that will never be used in practice 😭

SLOW DOWN, I can’t spend all of these tokens quick enough and I refuse to use a /goal based approach 😭

@notKartikk @magic_bubblez I like it, spy x family?

@magic_bubblez isn't it peak?

I am a very serious developer btw

@ChrisIzatt Dude you crushed this, even the simple things you baked into it!

Yo nobody ever said building a multi-provider ensemble harness in React without just forking another project (like the Codex app) would be easy but damnnnnn… Making great progress now though with releases available 👀👻 github.com/boggspa/TaskWr… github.com/boggspa/TaskWr…

@ChrisIzatt Valid point, but debugging is the fun part. Nice app though, I dig it bro 😎

@BlockedPaths Haha trying to although you probably know and understand the debugging and patience you gotta have to make sure things play ball in an equal/parity way otherwise if someone favours Grok or local models or something I don’t want them to have a sub-par experience haha 😭🤣

I've fallen victim to the loopz