Comment by SOLAR_FIELDS
15 days ago
https://github.com/anthropics/claude-code/issues/8477
https://github.com/anthropics/claude-code/issues/15263
https://github.com/anthropics/claude-code/issues/9099
https://github.com/anthropics/claude-code/issues/8371
It's very clear that Anthropic doesn't really want to expose the secret sauce to end users. I have to patch Claude every release to bring this functionality back.
I just assume that they realized that they can split the offering, and to charge for the top tier more. (Yes, even more.)
If Claude Code can replace an engineer, it should cost just a bit less than an engineer, not half as much.
Remember there are no moats in this industry - if anything one company might have a 2 month lead, sometimes. We've also noticed that companies paying OpenAI may swiftly shift to paying Google or Anthropic in a heartbeat.
That means the pricing is going to be competitive. You may still get your wish though, but instead of the price of an engineer remaining the same, it will cut itself down by 95%.
But then you pay for the less outrageously subsidized rates of API instead of the a bit less incredibly generous prices of the subscription.
Its not subsidized, in fact, they probably have very healthy margins on Claude Code.
18 replies →
I don't know about you, but I benefit so much from using Claude at work that I would gladly pay $80,000-$120,000 per year to keep using it.
Why would you gladly pay more than what it's worth? It's not an engineer you are hiring, it's AI. The whole point of it was to make intelligent workflows cheaper. If it's going to cost as much as an engineer, hire the engineer, at least you'd have an escape goat when things invariably go wrong.
10 replies →
What do you use it for, do you have example? For you to be ok with paying 80k to 120k I'm guessing its making you 375-450k a year?
1 reply →
Oh come on. That pays for more than 10 fte in some countries
13 replies →
that means customers will pay minimum 2x that much I think
[flagged]
Patching's not long for this world; Claude Code has moved to binary releases. Soon, the NPM release will just be a thin wrapper around the binary.
the binary is just bun. I wrote this to inspect CC
https://github.com/shepherdjerred/monorepo/tree/main/package...
Kinda crazy to buy a Javascript engine company to obfuscate your React TUI.
1 reply →
Where there's a will, there's a way
> It's very clear that Anthropic doesn't really want to expose the secret sauce to end users
Meanwhile, I am observing precisely how VS+Copilot works in my OAI logs with zero friction. Plug in your own API key and you can MITM everything via the provider's logging features.
> Plug in your own API key
I checked with ccusage (a cli tool that checks how much your Claude subscription tokens would have cost via the API).
My $200 a month subscription would have cost me more than $3000. The highest single day would have cost more than $300.
Gemini is cheaper, but not by much.
> to end users
To other actors who want to train a distilled version of Claude, more likely.
Honestly, just use OpenCode. It works with Claude Code Max, and the TUI is 100x better. The only thing that sucks is Compaction.
How much longer is Anthropic going to allow OpenCode to use Pro/Max subscriptions? Yes, it's technically possible, but it's against Anthropic's ToS. [1]
1: https://blog.devgenius.io/you-might-be-breaking-claudes-tos-...
Consider switching to an OpenAI subscription, which allows OpenCode use.
1 reply →
Doesn’t Claude code have an agents sdk that officially allows you to use the good parts?
1 reply →
There are also Azure versions of Opus
I have been unable to use OpenCode with my Claude Max subscription. It worked for awhile, but then it seems like Anthropic started blocking it.
What’s 100x better about the TUI?
Nope, OpenCode is nowhere near Claude Code.
It's amazing how much other agentic tools suck in comparison to Claude Code. I'd love to have a proper alternative. But they all suck. I keep trying them every few months and keep running back to Claude Code.
Just yesterday I installed Cursor and Codex, and removed both after a few hours.
Cursor disrespected my setting to ask before editing files. Codex renamed my tabs after I had named them. It also went ahead and edited a bunch of my files after a fresh install without asking me. The heck, the default behavior should have been to seek permission at least the first time.
OpenCode does not allow me to scrollback and edit a prior prompt for reuse. It also keeps throwing up all kinds of weird errors, especially when I'm trying to use free or lower cost models.
Gemini CLI reads strange Python files when I'm working on a Node.js project, what the heck. It also never fixed the diff display issues in the terminal; It's always so difficult for me to actually see what edits it is actually trying to make before it makes it. It also frequently throws random internal errors.
At this point, I'm not sure we'll be seeing a proper competitor to Claude Code anytime soon.
Hmmm, I used OpenCode for awhile and didn't have this experience. I felt like OpenCode was the better experience.
1 reply →
I use Opencode as my main driver, and I don’t experience what you have experienced.
For instance, opencode has /undo command which allows you to scroll back and edit a prior prompt. It also support forking conversations based on any prior message.
I think it depends on the set up. I overwrote the default planning agent prompt of opencode to fit my own use cases and my own mcp servers. I’ve been using OpenAI’s gpt codex models and they have been performing very well and I am able to make it do exactly what I ask it to do.
Claude code may do stuff fast, but in terms of quality and the ability to edit only what I want it to do, I don’t think it’s the best. Claude code often take shortcuts or do extra stuff that I didn’t ask.
5.3 Codex on cursor is better than Claude code
1 reply →
If they cared about that, they wouldn't expose the thinking blocks to the end-user client in the first place; they'd have the user-side context store hashes to the blocks (stored server-side) instead.
I don't suppose you could share a little on that patching process?
More likely 99.9% of users never press ctrl+o to see the thinking, so they don't consider it important enough to make a setting out of.
To be fair they have like 10,000 open issues / spam issues, it's probably insane out there for them to filter all of it haha
GitHub Issues as a customer support funnel is horrible. It's easy for them, but it hides all the important bugs and only surfaces "wanted features" that are thumbs-up'd alot. So you see "Highlight text X" as the top requested feature; meanwhile, 10% of users experience a critical bug, but they don't all find "the github issue" one user poorly wrote about it, so it has like 7 upvotes.
GitHub Codespaces has a critical bug that makes the copilot terminal integration unusable after 1 prompt, but the company has no idea, because there is no clear way to report it from the product, no customer support funnel, etc. There's 10 upvotes on a poorly-written sorta-related GH issue and no company response. People are paying for this feature and it's just broken.
Humans don't look at these anymore, Claude itself does. They've even said so.
Maybe they can use AI to figure out which ones are actually useful and which ones are not.
I thought the source code for the actual CLI was closed source. How are you patching it?
Claude code can reverse engineer it to a degree. Doing it for more than a single version is a PITA though. Easier to build you own client over their SDK.
I think it's more classic enshittification. Currently, as a percentage, still not many devs use it. In a few months or 1-2 years all these products will start to cater to the median developer and start to get dumbed down.