this post was submitted on 25 Dec 2025
53 points (98.2% liked)
ADHD
12391 readers
139 users here now
A casual community for people with ADHD
Values:
Acceptance, Openness, Understanding, Equality, Reciprocity.
Rules:
- No abusive, derogatory, or offensive post/comments.
- No porn, gore, spam, or advertisements allowed.
- Do not request for donations.
- Do not link to other social media or paywalled content.
- Do not gatekeep or diagnose.
- Mark NSFW content accordingly.
- No racism, homophobia, sexism, ableism, or ageism.
- Respectful venting, including dealing with oppressive neurotypical culture, is okay.
- Discussing other neurological problems like autism, anxiety, ptsd, and brain injury are allowed.
- Discussions regarding medication are allowed as long as you are describing your own situation and not telling others what to do (only qualified medical practitioners can prescribe medication).
Encouraged:
- Funny memes.
- Welcoming and accepting attitudes.
- Questions on confusing situations.
- Seeking and sharing support.
- Engagement in our values.
Relevant Lemmy communities:
lemmy.world/c/adhd will happily promote other ND communities as long as said communities demonstrate that they share our values.
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I was dabbling in local llms recently using Ollama to generate stories from prompts. It's fine but not something I'd consider something original like how I'd write.
But I guess it's better than nothing if I get stuck.
Ollama is really terrible, especially since it uses chat mode by default and a 2048 context.
If you run a better pretrain with a “notepad” UI like Mikupad, it’s like night and day. It follows your writing style because that’s the context it has to go on.
If you’re interested, tell me your hardware config and I can recommend something specific, but generally you’re going to want to run ik_llama.cpp with a big MoE base model, like GLM Air base. Use something like Q6/Q5.1 cache quantization, enable the hadamard option, and then tune the GPU layer count until it fills your vram.
Well of course. If you give it a short prompt, and it generates a story, that isn't going to be anything of value. But if you give it long prompts and have it give you ten different sets of three sentences that could follow, you have a goid shot that one of those either fits what you were thinking but couldn't get in words, or will stimulate a better sentence in your mind. It can be a block breaker. Don't ask it for whole stories, just sentences or paragraphs. Or even just to reword some thing you wrote if you don't like how it flowed.