Bruno Neto

@thsottiaux We need multi windows/sessions side by side please

I don't know why we ship on Saturdays now, but here are a bunch of nice improvements for Codex. Didn't feel right to hold this back until Tuesday when it all makes it a lot more delightful to use.

@bridgemindai Something is wrong with Claude again Hitting 5 hour limit easy easy Alreayd at 58% weekly on my 20x and 29% on my second 5x

Kimi K2.6 is 6x cheaper per token than Claude Opus 4.7. But per task? It's only 39% cheaper. $0.76 per task for Kimi K2.6. $1.24 for Claude Opus 4.7. Kimi burns so many tokens to complete a task that the 6x pricing advantage nearly disappears. Cheaper per token does not mean cheaper to use. If a model takes 2x the tokens and 7x longer to finish, the savings are an illusion. Stop comparing token prices. Compare cost per task.

@bridgemindai Guys, am I the only one who feels Codex too slow? Like, even on fast mode, it’s super slow

Claude Code just raised weekly limits by 50%. This puts Claude Code rate limits on par with Codex. I cancelled my Max plan twice over rate limits. SpaceXAI deal doubled the five hour limits. Removed peak hour limits. Now weekly limits up 50%. Three rate limit increases in two weeks. Anthropic is not playing around anymore. Claude Code with Claude Opus 4.7 now has unlimited limits.

@kyle_e_walker @duckdb Hi Kyle! Very nice! How did you access to the parcels? Regrid? I’ll in utility construction, do you know how I could get gis data for utility poles? Thanks. Your work is very inspiring

14.3 million parcels visualized across Texas. Draw a shape, query @duckdb, tabulate surface ownership instantly. Modern geospatial tooling brings powerful land analytics to your laptop.

@bridgemindai How did it generated those 20k?

My second NVIDIA DGX Spark just arrived. Connecting them today. Larger models tomorrow. My Hermes Agent has now generated over $20k. I am going to be running Nemotron 3 Super. Claude Opus 4.7 gets more expensive every week. Rate limits are higher than ever. I'm building an escape hatch. Stream tomorrow.

@bridgemindai Guys, how do I code with this? OpenCode? Anyone can help? Thanks

Kimi K2.6 is now live on OpenRouter. $0.95 per million input tokens. $4 per million output. 262K context window. Claude Opus 4.7 is $5 input, $25 output. That's 94% cheaper input. 95% cheaper output. And Kimi K2.6 beats Opus 4.6 on SWE-Bench Pro. The math doesn't lie.

@theo Can I still use 4.5? If yes, how? Thanks

Is Opus 4.7 the best model from Anthropic? No, that’s Mythos. Is it the best model we can use for code? No, that’s GPT-5.4. Is it the best model in the Opus line? No, that’s 4.5 It’s the best model released today I guess?

@bcherny I was anxiously waiting for that auto mode. Tks

We're also rolling out auto mode for Max users. This means no more permission prompts — give Claude a task, let it run, come back to verified work.

Opus 4.7 is in Claude Code today. It's more agentic, more precise, and a lot better at long-running work. It carries context across sessions and handles ambiguity much better.

@theo Agreed

I feel bad dunking on them so much but it's genuinely absurd how bad the new Claude Code desktop app is. You can feel the vibe code leaking everywhere. Every "feature" is barely integrated and full of edge cases that weren't considered. Every menu feels barren, stuffed in last second for some random toggle. Every hotkey breaks as soon as you try to do anything else. I've lost track of how many bugs I've encountered. I found at least 40 in under an hour. And it's all truly absurd arcane shit. Stuff like voice mode typing in all input boxes instead of just the one you have focused. Any one of these issues would have been enough for me to do a massive post-mortem and likely fire someone. A $400b company shipping this is absurd. I feel like I'm going mad. How does anyone seriously use this?? It is broken on fundamental levels that are hard to comprehend. How are we supposed to trust the code these models produce if Anthropic's official showcases are absolute slop? Dedicated video on this coming tomorrow. Just needed to get this off my chest.

@bridgemindai have you notice any changes in Opus 1M context? When they first released I don't remember having to compact it at all. Now, it feels like the regular old context window @claudeai

@bridgemindai Same here. Frustrating.

Claude Code is completely down right now. 500 internal server errors. Not rate limits. The servers are crashing. Everyone is getting hit. Middle of a Monday. Peak hours. $200/month Max plan. 529 errors last week. 401 auth errors the week before. Now 500s. This is a different error every week. Fix this @claudeai

@amorriscode @rileybrown I tried it, but the extension in vs code is a lot better

@rileybrown New CC desktop version comes out next week. It’s a lot better.

Codex App > Claude Desktop App

@amorriscode @rileybrown Still very laggy

@ShaneMac @openclaw Loved it! Here I’m to test it

Apps are dead. We just shipped this. Screenshot anything in the App Store → drop it into chat → swipe up for an instant @openclaw agent → delete app, save $29/year Add agents to every group or chat. No signup. No download. New agent, every group. Invite only for now.

@MaziyarPanahi @Prince_Canuma How is the camera setup to the macbook? Is it live or a recording?

Gemma 4 sees a kid and three dogs. Decides what matters. Calls SAM 3.1 Mask and bounding box. Spotlight on subjects. Background blur. Background pixelation. Four effects. Fully agentic. Two models talking to each other on a MacBook. No App. No cloud. What would you edit?

@MaziyarPanahi @Prince_Canuma Loved this thing. Trying to find where to use it. Lol

Me trying to learn Claude Code to avoid losing my job, when there are much bigger issues in 2026

O melhor lugar do planeta pra se comer bem se chama Minas Gerais.

@tarredev Of course not lol

@brunomnetoo And has it already been approved by the Apple Store? That’s the worse part

Acabo de enviar mi primera app a la App Store. Aquí va lo que nadie te cuenta antes de hacerlo 🧵