Oh I do. Headless 128GB RAM machine serving llama.cpp with a number of local models that I use on a daily basis.
• Qwen3-VL picks up new images in a NAS, auto captions and adds the text descriptions as a hidden EXIF layer into the image, which is used for fast search and organization in conjunction with a Qdrant vector database.
• Gemma3:27b is used for personal translation work (mostly English and Chinese).
• Some small 8b models (like llama3.1) for sentiment analysis on text.
But haven't really tried using local LLMs in conjunction with agentic harnesses yet.
My secret sauce is to use LLAMAcpp's reasoning-budget and reasoning-message that trigger cut off to overthinking with a message that says to either us subagents or compress the context. opencode's dynamic context pruning plugin can get you pretty far into the stratosphere.
> The Sleev (the project has been renamed to make a startup) creator was shilling their project in the OpenCode Discord. That person is very convinced they have something that no one has ever built before. They focused on token reduction without any real evals for capability impacts.
I'm generally against this context pruning without prompting or details. Sleev is very opaque about how it works and definitely will bust your cache.
No. the opencode plugin dynamic context pruning is essentially just labeling the context and then summarizing it; you can manually expand the context if you missed something, but most of the time, I'm using it to extend context into different scopes rather than trying to achieve the same goal.
There's a few times it gets too dumb to cope, but in comparison with other coding agents, it seems smarter.
> My secret sauce is to use LLAMAcpp's reasoning-budget and reasoning-message that trigger cut off to overthinking with a message that says to either us subagents or compress the context
Thanks for the tip - I like this a lot. I remember having to do a lot of tweaking to curtail Qwen QwQ-32b when it would go down an endless psychotic recursive reasoning loops as part of its "chain of reasoning."
• Qwen3-VL picks up new images in a NAS, auto captions and adds the text descriptions as a hidden EXIF layer into the image, which is used for fast search and organization in conjunction with a Qdrant vector database.
• Gemma3:27b is used for personal translation work (mostly English and Chinese).
• Some small 8b models (like llama3.1) for sentiment analysis on text.
But haven't really tried using local LLMs in conjunction with agentic harnesses yet.