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

Claude Sonnet 4.5

anthropic.com

Post reply on HN
Print view View original

91–100 of 819 posts

Re: Claude Sonnet 4.5

#91

Post by punkpeye »

cryptoz wrote: post #4

I've really got to refactor my side project which I tailored to just use OpenAI API calls. I think the Anthropic APIs are a bit different so I just never put in the energy to support the changes. I think I remember reading that there are tools to simpify this kind of work, to support multiple LLM APIs? I'm sure I could do it manually but how do you all support multiple API providers that have some differences in the…

OpenRouter, Glama ( https://glama.ai/gateway/models/claude-sonnet-4-5-20250929 ), AWS Bedrock, all of them provide you access to all of the AI models via OpenAI compatible API.

Re: Claude Sonnet 4.5

#92

Post by johnfn »

blurbleblurble wrote: post #56
yewenjie wrote: post #5

Looking at the chart here, it seems like Sonnet 4 was already better than GPT-5-codex in the SWE verified benchmark. However, my subjective personal experience was GPT-5-codex was far better at complex problems than Claude Code.

GPT-5 is like the guy on the baseball team that's really good at hitting home runs but can't do basic shit in the outfield. It also consistently gets into drama with the other agents e.g. the other day when I told it we were switching to claude code for executing changes, after badmouthing claude's entirely reasonable and measured analysis it went ahead and decided to `git reset --hard` even after I twice pushed back…

That’s such a great analogy. I always say GPT is like the genius that completely lacks common sense. One of my favorite things is when I asked it why the WiFi wasn’t working, and showed it a photo of our wiring. It said that I should tell support:

> “My media panel has a Cat6 patch panel but no visible ONT or labeled RJ45 hand-off. Please locate/activate the Ethernet hand-off for my unit and tell me which jack in the panel is the feed so I can patch it to the Living Room.”

Really, GPT? Not just “can you set up the WiFi”??!

Re: Claude Sonnet 4.5

#93

Post by renewiltord »

cellis wrote: post #21
yewenjie wrote: post #5

Looking at the chart here, it seems like Sonnet 4 was already better than GPT-5-codex in the SWE verified benchmark. However, my subjective personal experience was GPT-5-codex was far better at complex problems than Claude Code.

Opposite for me…5-codex high ran out of tokens extremely quickly and didn’t adhere as well to the agents.md as Claude did to the Claude.md, perhaps because it insists on writing extremely complicated bash scripts or whole python programs to execute what should be simple commands.

gpt-5 command line use is bizarre. It always writes extraordinarily complicated pipelines that Claude instead just writes simple commands for.

My use case does better with the latter because frequently the agent fails to do things and then can't look back at intermediate.

E.g. Command | Complicated Grep | Complicated Sed

Is way worse than multistep

Command > tmpfile

And then grep etc. Because latter can reuse tmpfile if grep is wrong.

Re: Claude Sonnet 4.5

#94

Post by XMPPwocky »

zurfer wrote: post #31

Same price and a 4.5 bp jump from 72.7 to 77.2 SWEBench Pretty solid progress for roughly 4 months.

nit: assuming you mean basis points, one basis point is 0.01%. 4.5bp would be 72.7% to 72.71%. this is 450bp!

Re: Claude Sonnet 4.5

#95

Post by doctoboggan »

vrosas wrote: post #86
blurbleblurble wrote: post #56

Earlier quoted context omitted.

GPT-5 is like the guy on the baseball team that's really good at hitting home runs but can't do basic shit in the outfield. It also consistently gets into drama with the other agents e.g. the other day when I told it we were switching to claude code for executing changes, after badmouthing claude's entirely reasonable and measured analysis it went ahead and decided to `git reset --hard` even after I twice pushed back…

Why are you having a conversation with your LLM about other agents?

I do it as well. I have a Claude code instance running in my backend repo, and one running in my frontend repo. If there is required coordination, I have the backend agent write a report for the front end agent about the new backend capabilities, or have the front end agent write a report requesting a new endpoint that would simplify the code.

Lots of other people also follow the architect and builder pattern, where one agent architects the feature while the other agent does the actual implementation.

Re: Claude Sonnet 4.5

#96

Post by wohoef »

zurfer wrote: post #57
zurfer wrote: post #31

Same price and a 4.5 bp jump from 72.7 to 77.2 SWEBench Pretty solid progress for roughly 4 months.

Also getting a perfect score on AIME (math) is pretty cool. Tongue in cheek: if we progress linearly from here software engineering as defined by SWE bench is solved in 23 months.

Just a few months ago people were still talking about exponential progress. The fact that we’re already going for just linear progress is not a good sign

Re: Claude Sonnet 4.5

#97

Post by Kuinox »

AlecSchueler wrote: post #82
eli wrote: post #66

Earlier quoted context omitted.

Assuming they updated the crawled training data, just having a bunch of examples of specifically pelicans on bicycles from other models is likely to make a difference.

But then how does the quality increase? Normally we hear that when models are trained on the output of other models the style becomes very muted and various other issues start to appear. But this probably the best pelicans on a bicycle I've ever seen, by quite some margin.

Just compare it with a human on a bicycle, you would see that LLMs are weirdly good at drawing pelicans in SVG but not humans.

Re: Claude Sonnet 4.5

#98

Post by MichealCodes »

rudedogg wrote: post #70

I just ran this through a simple change I’ve asked Sonnet 4 and Opus 4.1, and it fails too. It’s a simple substitution request where I provide a Lint error that suggests the correct change. All the models fail. I could ask someone with no development experience to do this change and they could. I worry everyone is chasing benchmarks to the detriment of general performance. Or the next token weight for the incorrect c…

More like churning benchmarks... Release new model at max power, get all the benchmark glory, silently reduce model capability in the following weeks, repeat by releasing newer, smarter model.

Re: Claude Sonnet 4.5

#99

Post by doctoboggan »

ancorevard wrote: post #88

Can't use Anthropic models in Cursor. Completely cost prohibitive compared to gpt-5 and grok models. Why is this? Does Anthropic have just higher infrastructure costs compared to OpenAI/xAI?

Possibly, or they are pricing for sustainability and OpenAI/xAI are just burning through VC money.

Re: Claude Sonnet 4.5

#100

Post by tux3 »

blurbleblurble wrote: post #56
yewenjie wrote: post #5

Looking at the chart here, it seems like Sonnet 4 was already better than GPT-5-codex in the SWE verified benchmark. However, my subjective personal experience was GPT-5-codex was far better at complex problems than Claude Code.

GPT-5 is like the guy on the baseball team that's really good at hitting home runs but can't do basic shit in the outfield. It also consistently gets into drama with the other agents e.g. the other day when I told it we were switching to claude code for executing changes, after badmouthing claude's entirely reasonable and measured analysis it went ahead and decided to `git reset --hard` even after I twice pushed back…

That's great given that the goal of OAI is to train artificial superintelligence first, hoping that the previous version of the AI will help us control the bigger AI.

If GPT-5 is learning to fight and undo other models, we're in for a bright future. Twice as bright.

Post reply on HN