Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A week of using Codex more than Claude

allaboutcoding.ghinda.com

Post reply on HN
Print view View original

281–290 of 311 posts

Re: A week of using Codex more than Claude

#281

Post by thefourthchime »

timr wrote: post #140
seanmcdirmid wrote: post #139

Earlier quoted context omitted.

If politics isn’t a concern, why not just use deepseek?

I am trying all of them. At this time, for me, Grok hits the sweet spot of quality, speed, cost and ease of use. I’m absolutely “hot money” when it comes to coding models. These things are commodities.

I recently tried a Cursor ultra and Grokbot. Holy shit, dude.

Half the coding I do is for my phone now because Cursor Ultra agents have their own VMs that are spun up specifically for each project.

Grok bot has a bunch of agents that'll share a VM and they can do pretty much anything you can do digitally. Right on I have them checking slick deals every morning for a pellet smoker.

I had it book a date night for me. I had it fix one of my projects by rebuilding my website and republishing it and then checking one of the container runs to see if it has errors on it.

I had a call different banks to figure out which phone navigation tree to get through and put someone on the phone for me, and then call me

The list just goes on and on.

Re: A week of using Codex more than Claude

#282

Post by thefourthchime »

mediaman wrote: post #84
trjordan wrote: post #31

Earlier quoted context omitted.

Not mentioning Grok 4.6 here is a crime. Fast and accurate. And it can communicate, unlike the gobbledygook that comes out of Claude.

As a coda to this, anyone using grok 4.6 via API pricing should be aware that while their headline pricing is good, the pricing that actually matters is pretty bad. Their cache read costs are $0.50 per million, or 25% of the cost of uncached reads. The industry standard is a 90% discount, so cache costs you 10% of uncached. So that means 5.6 Sol actually costs less per million cache reads - $0.40/million. If you are…

I’ll agree with this. I liked grok build, but cost-wise, it’s just not competitive with cursor and codex

Personally, I’ve switched to cursor ultra, which picks between about five models to do whatever you want.

It's weird not to pick the best model all the time, if you can. But I got so frustrated with GPT-5.6 spending forever and then doing the wrong thing and making bugs.

I'd rather have auto do the wrong thing fast and make bugs and then it can fix them. It's a trade-off, but I found the speed better. And you can always switch to a better model if you don't trust It.

Re: A week of using Codex more than Claude

#283

Post by zarzavat »

oblio wrote: post #240
zarzavat wrote: post #188

Earlier quoted context omitted.

You may be causing a lot of cache misses. You have to use the caching efficiently otherwise you can burn up any plan in any amount of time.

How do you use the cache efficiently?

You have to keep your session warm in cache. Keep the AI talking/thinking. If you have not touched a session for few minutes then /clear and start a new session.

Providers will generally keep your session in cache for at least 5 minutes, possibly hours. The exact cache policy depends on the provider.

If your session expires from cache then the next time you send a message you will have to pay for all the tokens you had used in context up until that point again. e.g. if you have 200k tokens in context then if your session goes cold and you send a message after expiry you will have to pay for those 200k tokens again.

With 1M contexts especially you have to be extremely careful that you don't end up resubmitting requests for hundreds of thousands of tokens again and again.

Try to get yourself and the model to use disk for medium-term context rather than model context, that way it's much easier to /clear and restart if you need to go to the bathroom or something.

Re: A week of using Codex more than Claude

#284

Post by tnchr »

dbbk wrote: post #274

I find myself in an awkward tug of war between Claude and Codex. Claude is the buggiest product, always breaking in bizarre and aggravating ways. Codex is much more polished, especially on mobile. However the main thing that's prevented me from adopting Codex full time is its poor implementation of plan mode. While Claude persists it to disk and re-reads it in full on compaction, Codex's just lives in context, so it…

I ask it just to make a plan file

Re: A week of using Codex more than Claude

#286

Post by skeledrew »

iammrpayments wrote: post #193
skeledrew wrote: post #119

Earlier quoted context omitted.

Confusion usually surfaces immediately, so it can be steered properly. Being sure of a decision that turns out to be wrong usually means a bug that could bite when you least expect it.

No it doesn’t. The comments end up complicating simple tasks, and if you open the repo after a week it takes a while to find out why Claude is reading irrelevant files or replying about irrelevant things

If that happens you do notice that something is off though, and can steer accordingly. Unless you aren't monitoring the agent, which I suspect is the real issue here; people send a prompt, go do whatever else, return and send the next prompt or use/release the artifact, without reviewing the transcript to see how Claude - or whatever agent - did.

Re: A week of using Codex more than Claude

#287

Post by Makson179 »

vardalab wrote: post #179
c0rruptbytes wrote: post #117

Earlier quoted context omitted.

Pi by itself is more than capable, OMP is okay but you really don't need much for a great harness (these models are RL trained to hell to be a coding agent, sometimes less is more) I run a lot of SlopCodeBench - https://github.com/michaelasper/benchmarks Fable/Sol/GLM 5.3/Kimi are its league (in that order) Deepseek/Opus is solid Qwen 27B is the floor - there's no reason to use Sonnet/Terra/Haiku For everyday activit…

Using Sol XHigh or even High will deplete the Pro sub pretty fast in my experience if one is running any sort of automations in their harnesses. Sol-medium lets me squeek by with it using lesser subagents. Using ninfer on 5090 and 35BA3B qwen 3.6 is also kind of cool to get a local cerebras experience at 600 tk/, it does make errors so 27B is actually faster at the end at 140-150 tk/s. 35B is great though at digging…

Effort level might not be the root of the problem. In my runs reasoning is around 10% of the cost and writing code maybe 20%. The rest is 5.6 re-reading files. Model itself became way too meticulous.

Re: A week of using Codex more than Claude

#288

Post by skeledrew »

fooster wrote: post #199
skeledrew wrote: post #113

Earlier quoted context omitted.

It does visibly use a few more tokens in the short term. But it hurts Claude's ability in the long term if removed because it has to either think more every time it encounters that section in the code (more invisible thinking tokens used), or there's an increased risk of making a maladaptive change without context. Heck I've been bitten enough times already by it, though more recently it's cross project concern where…

No it doesn’t. The comments are hugely verbose trash that immediately becomes outdated and brittle and confusing for humans and Claude alike.

If that were the case, the habit would have been trained out of it by the LLM researchers at Anthropic, etc. As for humans... well Idk about you but I'm not reading every line of generated code. I read the live transcript, and only dive in if something is off that the agent isn't resolving.

Re: A week of using Codex more than Claude

#289

Post by skeledrew »

grinco wrote: post #249
skeledrew wrote: post #113

Earlier quoted context omitted.

It does visibly use a few more tokens in the short term. But it hurts Claude's ability in the long term if removed because it has to either think more every time it encounters that section in the code (more invisible thinking tokens used), or there's an increased risk of making a maladaptive change without context. Heck I've been bitten enough times already by it, though more recently it's cross project concern where…

What you need is RAG and better context management. I use vornik (it has plugins for both codex and Claude) that gives both the long term memory you mention. Hope this helps.

Sounds like a bunch of extra tool calls, and dependence on a hopefully accurate retrieval system. Vs the extra context just being available where it's relevant.

Re: A week of using Codex more than Claude

#290

Post by fooster »

skeledrew wrote: post #288
fooster wrote: post #199

Earlier quoted context omitted.

No it doesn’t. The comments are hugely verbose trash that immediately becomes outdated and brittle and confusing for humans and Claude alike.

If that were the case, the habit would have been trained out of it by the LLM researchers at Anthropic, etc. As for humans... well Idk about you but I'm not reading every line of generated code. I read the live transcript, and only dive in if something is off that the agent isn't resolving.

I'm certainly reading the code and comments, and I encourage all of my team members to do so. You are assuming that this excessive and nonsensnical comment choice is deliberate. It cannot be, or should not be because it regularly confused our team, and it regularly confuses claude also.

Also many many people keep saying the same thing and you keep repeating adnausium the same tired comments. I get it, you think the comments are great and are valuable.

Read the entire thread https://news.ycombinator.com/item?id=49393378

The concensus in this thread is that for multiple reasons the excessive comments are in fact actively harmful. The listed reasons are:

1. Document the conversation, not the code — narrate the back-and-forth that produced the change rather than what the code does

2. Reference intermediate states that never shipped ("previously X, now Y")

3. Cite plan documents and session artifacts (`per section 2.1 of addition-implementation-plan.md`) meaningless to a future reader

4. Belong in commit messages or PR comments, not source files

5. Go stale immediately — describe a state the code is no longer in

6. Launder incorrect code as intentional, making bugs harder to spot

7. Build a false Chesterton's fence around mistakes

8. Use defensive prose ("this is not cosmetic", "prevents the critical bug that shipped once") that asserts importance instead of conveying information

9. Describe what the code doesn't do — relevant in the moment, not in the codebase

10. Confuse later agent sessions, sending them to read irrelevant files

11. Consume context tokens on every read

12. Force reviewers to manually delete the litter

13. Resist correction — telling Claude to be concise doesn't durably stick

Please consider that your opinion may need to adjusted.

Post reply on HN