Every time I try Opus 4.1 in the Claude client, it just breaks – goes around in circles even for relatively simple tasks and then throws up an error. I'm sure it is not that Opus is bad, but it seems to have outgrown its Claude interface shackles
@petergostev
-
GPT-5 Pro capabilities: unsolved problems and Project Euler challenges
By
–
Elliot are there any problems worth trying with GPT-5 Pro do you think? Something that GPT-5 / others haven't been able to solve? Looks like Project Euler solutions are hidden
-
Anthropic reasoning traces differ from OpenAI DeepSeek approaches
By
–
When you look at the reasoning traces of Anthropic models, it just reads to me very unlike the OpenAI / DeepSeek reasoning. Anthropic looks a lot more like planning rather than eg deliberating on the answer, backtracking etc – like OpenAI and DeepSeek do, feels different
-
Opus 4.1 costs 3x more than GPT-5 and Gemini 2.5
By
–
I ran a semi-scientific cost test of 3x top models: Opus 4.1 was 3x the cost of GPT-5 and Gemini 2.5 Pro on real tasks. I gave the same prompt to each model (with reasoning enabled) and looked at the cost, without normalising for tokens. The point was to capture real cost and
-
GPT-5 Pro Evaluates Situational Awareness Claims Accuracy
By
–
I got GPT-5 Pro to research situational awareness by @leopoldasch and assess whether his claims are True, Tracking True, Tracking False, False or Not Enough Info. So far it is doing pretty well 8 out of 13 True or Tracking True, only 1 tracking false. Including the link below
-
GPT-4o-mini Token Usage Spike: Large User Activity Theory
By
–
If you go back further you can see that gpt-4o-mini tokens went up for a couple of months and then went down, my guess is that it was just someone with big usage milking gpt-4o-mini. Doubt this actually means anything
-
GPT-5 Thinking dominates web development benchmarks with 65% win rate
By
–
GPT-5 (Thinking) tops the Web Dev @lmarena_ai
, winning ~65% of matchups and losing ~20%. This benchmark focuses on front-end development, fairly simple apps, so it is not representative of all coding. But OpenAI have made huge strides in front-end, given that their previous best -
GPT-5 Pro Competence Compared to o1 and o3 Models
By
–
GPT-5 Pro is under-hyped. Pretty much every time I try it, I’m surprised by how competent and coherent the response is. – o1-pro was an incredible model, way ahead of its time, way better than o1 – o3 was better because of its search – o3-pro was a little disappointing