Mr Tactical

@droidbuilds Codex has no opinion or view about app development decisions. It does as it’s told, that’s it. Claude suggests improvements without you having to ask Claude to suggest improvements. Claude is like a 2nd developer rather than a tool.

what’s stopping you from switching to Codex?

@dreamsofcode_io Not much of a joy when you watch Codex or Claude Code at work… unmatchable speed…

Writing code by hand is a joy. Would recommend. 10/10

@theo @MiniMax_AI @datacurve 👀 📈

@MiniMax_AI @datacurve you know the drill

Introducing MiniMax M3: The First Open-Weights Model to Combine Three Frontier Capabilities - Coding & Agentic Frontier: 59.0% SWE-Bench Pro, 66.0% Terminal Bench 2.1, 34.8% SWE-fficiency, 28.8% KernelBench Hard, 74.2% MCP Atlas - MiniMax Sparse Attention scales context to 1M - Natively Multimodal from Step Zero API: platform.minimax.io Token Plan: platform.minimax.io/subscribe/toke… 🚀New! MiniMax Code: code.minimax.io Weights & Tech Report in ~10 Days

@ClaudeCodeLog 👀

Claude Code 2.1.159 is about to be released #cccnext

@kr0der EXACTLY - I have the same experience.

i’ve been testing Opus 4.8 for 1-2 days and i see why a lot of people use Claude Code Opus is insanely easy to talk to - it feels like a coworker who gets your intent and also suggests useful next tasks Codex is more of a robot that’s programmed to follow your instructions both are really good imo so use either one

@bridgemindai Mythos is gonna be a hell of a model 👀

I spent 2 billion tokens on Claude Opus 4.8 in 48 hours. Here's the honest truth. The good: most intelligent model in the world on Artificial Analysis. Best coding model available. Builds things GPT 5.5 and Opus 4.7 couldn't. The bad: this is not a leap. Same hallucination rate, same knowledge cutoff, same token usage as 4.7. Anthropic literally called it "modest but tangible." The ugly: I ran 105 UltraCode agents, shipped to production, and 4.8 introduced a bug. Then it failed to fix its own bug 8 times. Full rollback. These models design clean UIs and ship features. They still can't save you when things break. I'm waiting for the leap. Mythos in June. Full review below.

As an Apple developer: Most devs overhype raw benchmarks. I test models the only way that matters - in REAL Xcode projects. Claude consistently beats Codex for me on SwiftUI architecture and complex state management. Codex wins on quick edits where you just describe what you want. Real developers feel the difference in actual shipping speed. Who else is running both daily? What’s winning for you right now? #Xcode #ClaudeAI #iOSDev #AppleDeveloper #SwiftUI

@thegenioo You probably had an issue with your prompt - Claude has never broken any of my apps.

I hate Opus 4.8 (or claude overall), literally broke my app Codex limits got cooked so asked it to do a minor change and it fucked it all up

@nitishxyz @claudeai This never happens to me 🤷🏻‍♂️

opus 4.8 is the goat! ... of hallucinations.

@egorkabantsov App Dev with Xcode.

@TacticalDevX What are you using it for?

devs, vibe coders, founders which are y’all using for coding now? gpt 5.5 or opus 4.8

Are we getting an update for Codex Today? @thsottiaux #codex

@thsottiaux I use Both Codex and Claude.

I looked at a number today on a codex dashboard and it made me happy. More news about the number soon. 👀 Thanks to everyone who keeps adopting codex. We are still early. So early.

@thsottiaux I test the model myself

Do you still trust benchmarks or do you just listen to your friends? What makes you try a new model?

@thsottiaux 👀

The world is more fun with codex glasses

@themikejung Well actually your dm's are not open. 👀

I’m not a fan of these generic 'let's connect' posts, but here I am playing the algorithm game. 😅 I'm a developer building in the Web3/SaaS space and I'm looking to connect with actual builders: People who prioritize product over fluff Anyone tackling SaaS, Web tools, or Automation Those who are tired of the noise and just want to ship DMs are open. If you’re actually building something, let's connect and grow together. 🛠️ #BuildInPublic #SaaS #Web3 #IndieHackers #DevCommunity

@theillbo I’m in EU and I believe I just got it Today.

We will ever see Codex Computer use in EU/UK... 😭 Sooner or later?

Opus 4.8 drops and suddenly the same people who told you to switch to Codex 10 days ago are calling Claude Code a game-changer. I don’t move back and forth. I stick with @claudeai. Adaptability is good. Chasing every new shiny thing is just procrastination with extra steps.

@thsottiaux When GPT 5.6? @thsottiaux

Nobody knows Codex better than itself