@justbyte_ Yes, because to get any stretch of work done on your Claude sub you'll have to use Sonnet 90% of the time. So you're paying all that money to not get the best model.
With Codex you can yolo with 5.4 all day.
English
Jeff Scott
9.1K posts
Jeff Scott
@JeffScottDev
Canadian, Husband, Father, Software Engineer. Life is one big party when you're still young. Can't be a fool, son, what about the long run?
Canada Se unió Temmuz 2009
397 Siguiendo827 Seguidores
@Shpigford You know they are just restarting the process every time (we have all been there).
English
i don't believe this uptime for one second. not a day goes by that i don't hit 500 errors.
i'd be shocked if they truly even have ONE nine of uptime.
i get it. unprecedented scale, etc etc. but i'm struggling to find a service i trust less to actually work.
/end rant
Josh Pigford@Shpigford
i know i'm building more in a 24 hour period that i could artisanally build in a month, but the daily outages are truly maddening.
English
So I cancelled my @claudeai subscription but still have to suffer through the remaining usage that I paid for.
This is what I have done to actually get work done for more than an hour every 5 hours.
I use ONLY Sonnet 4.6 thinking High with dynamic reasoning disabled. I have been planning smaller more focused changes. Nothing broad or too ambitious.
Everyone once and awhile, like once a block, I have to switch to Opus to solve some deep technical blocker, but for the most part Sonnet can cook. Up until this outage I was at 87% with about an hour to go (not bad).
English
Good news everyone, Claude Code isn't down the status site says it's up!
This is just a figment of your imagination:
"API Error: 500 {"type":"error","error":{"type":"api_error","message":"Internal server
error"},"request_id":"req_XXX"} · check status.claude.com "
status.claude.com
English
@JeffScottDev Thanks. Weird thing, the Claude status website looks good. 🙄
English
@callstackio Once my app gets to a certain size and maturity it has Auth. I wish there was a pattern for deep linking with auth in these tools.
English
Introducing Agent React DevTools.
AI agents can now inspect React apps beyond the UI:
→ Component tree and state
→ Renders and performance
→ Structured output for debugging
Works with React and React Native.
Learn more ⬇️
English
Jeff Scott retuiteado
we just made pdf parsing better. again!
Firecrawl@firecrawl
Introducing Fire-PDF, our new Rust-based parsing engine 🔥 - Convert PDFs into markdown 5x faster - Extract full tables and preserve formulas - Zero config required
English
The future is a company of 25 hitters spending 25k/mo each, clearing 25m ARR/FTE
Eric Vishria@ericvishria
Seeing this all over. The efficiency and rationalization push is coming. Initially it was just getting everyone to use AI all over. "Opus everything!" Now the spend is material and companies will rationalize.
English
@YoavCodes The path is determined by whatever LLMs are currently good at writing.
English
Zig is the path.
Zig and Typescript will eat the world.
Viacheslav Biriukov@brk0v
⚡️ Zig 0.16 is out. And the new I/O model is a huge shift. • Swap implementations (threaded, evented, etc.) • Write code that looks blocking but runs async • Composable like allocators #toc-IO-as-an-Interface" target="_blank" rel="nofollow noopener">ziglang.org/download/0.16.…
#zig #ziglang English
@minchoi This might be the most unimpressive thing I have seen in a long time. Would have been cool 10 years ago maybe. Wheels? C'mon.
English
English
@JeffScottDev @cb_doge @bridgebench I have no idea what “elite coding model” means, but it sounds serious! 😅 Will Grok it. 😆
English
Grok 4.20 Reasoning just took the #1 spot on the BridgeBench reasoning benchmark. 🔥
Beating GPT-5.4, Claude Opus 4.6, Google Gemini and others.
Week after week, Grok keeps climbing across benchmarks. 🚀
English
English
I only use Grok. My life has had such an upgrade since Grok’s arrival! Nowadays I even teach others how to use it - and they are surprised how much you can do with it.
Yet, I am baffled that esp. academia doesn’t switch to Grok. Personally, I refuse to downgrade to other AIs. For that’s how I experience it - as a downgrade. Perhaps an idea to make colleges an offer they cannot refuse?
English
GLM-5.1 > Claude Code (Opus 4.6)?
I'm tripping or CC has become very bad but built a Three.js racing game to eval and it's extremely impressive. Thoughts:
- One-shot car physics with real drift mechanics (this is hard)
- My fav part: Awesome at self iterating (with no vision!) created 20+ Bun.WebView debugging tools to drive the car programmatically and read game state. Proved a winding bug with vector math without ever seeing the screen
- 531-line racing AI in a single write: 4 personalities, curvature map, racing lines, tactical drifting. Built telemetry tools to compare player vs AI speed curves and data-tuned parameters
- All assets from scratch: 3D models, procedural textures, sky shader, engine sounds, spatial AI audio!
- Can do hard math: proved road normals pointed DOWN via vector cross products, computed track curvature normalized by arc length to tune AI cornering speed
You are going to hear about this model a lot in the next months - open source let's go 🚀🚀
English
We just all need to realize that Enterprise customers are paying @AnthropicAI MUCH more money than we are; with built in profit margins.
Unfortunately I don't think they have any motivation to make our experience as developers any better, nor do they care if a percentage of us leave.
If I were to guess they are actually trying to drive attrition from the power users that are costing them money each month off the Max plans.
English