this post was submitted on 10 Aug 2025
98 points (99.0% liked)

technology

23929 readers
165 users here now

On the road to fully automated luxury gay space communism.

Spreading Linux propaganda since 2020

Rules:

founded 5 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
[–] jackmaoist@hexbear.net 14 points 2 weeks ago (5 children)

chatgpt is dogshit anyways and only surviving due to being the first and being free. So they have to burn money to stay relevant and hopefully not lose users to better models. GPT-5 is essentially a cost saving model and is the start of enshittification of the industry.

I use Claude for dev related stuff and it only provides limited queries so they can keep their model accurate since costs are limited.

Gemini already produces way better results than chatgpt ever did and is really good at research.

Perplexity can be a decent search engine.

Even i-am-adolf-hitler AI is better than chatgpt in most things although I'd rather not use it.

[–] invo_rt@hexbear.net 1 points 2 weeks ago (3 children)

What model do you recommend for brainstorming creative pursuits? As an adult with a "Serious Job™", it's hard for me to find anyone that has an interest in being creative. I find LLMs to give me enough feedback to help me flesh my ideas out.

[–] piccolo@hexbear.net 3 points 2 weeks ago (1 children)

For writing in English, GLM 4.5 is pretty good, open weights, and free or very cheap (free if you want just a chat interface, go to z.ai; cheap if you want API access, I'd recommend OpenRouter). It's imo the best non-closed source LLM for writing in English. DeepSeek can be good for that too, but I've found that it can sometimes produce sentences that flow a lot worse (you can use DeepSeek for free via their website as a chat interface). For other creative pursuits, I'm not sure - if you give me an idea of what you'd want out of one, I can try to give you advice.

[–] swelter_spark@reddthat.com 1 points 2 weeks ago

GLM is great. I liked 4 a lot, and 4.5 hasn't been a disappointment.

load more comments (1 replies)
load more comments (2 replies)