
metamike
3.4K posts

metamike
@itsmetamike
educator, consultant and proponent of the open metaverse
Los Angeles Katılım Haziran 2021
326 Takip Edilen2.3K Takipçiler
Sabitlenmiş Tweet

Eliza Agent Weaver
My @Gaianet_AI Autonomous Hackathon Submission
Eliza is a lightweight AI agent framework developed by @ai16zdao. It leverages Character files - JSON-formatted configurations that define an AI character's personality, knowledge, and behavior patterns.
Eliza Agent Weaver enables you to develop a set of Character files based on your own lore, and connects the narratives of multiple agents together through their character files.
If you want to support, please upvote my submission on @jokerace_io: jokerace.io/contest/base/0…
Github: github.com/itsmetamike/el…
English

@Teknium tbf I've never heard of NR before so I'm happy if openclaw gets proper competition, it pushes things forward
I'll try this, fine
English

@chrisflores_us I would love to hear more about where it’s falling short. Tbh didn’t think it would even be close to feasible to run the tool calling necessary for a successful Claude code session
English
metamike retweetledi

straight lines in log space don’t even get me high anymore. i need confidence intervals spanning an order of magnitude and maybe pointing to fledgling exponentials in log space just to feel alive
METR@METR_Evals
We estimate that Claude Opus 4.6 has a 50%-time-horizon of around 14.5 hours (95% CI of 6 hrs to 98 hrs) on software tasks. While this is the highest point estimate we’ve reported, this measurement is extremely noisy because our current task suite is nearly saturated.
English

I always wondered which Ollama model my machine could run properly.
Now you can easily do that using llms-checker
It's a CLI tool that checks your system specs and recommends the best models to run on Ollama based on what your machine
npm install -g llm-checker
github.com/Pavelevich/llm…
English

@sbknana @MatthewBerman Do these actually get the job done though?
English

@MatthewBerman Best local model if you've got 8-11GB VRAM:
qwen3:8b - best all-around, tool calling works
deepseek-r1:8b - reasoning and math
qwen2.5-coder:7b - pure coding
curl -fsSL ollama.ai/install.sh | sh
ollama pull qwen3:8b
English
metamike retweetledi

Me at every “2025 and the Impact of AI” meetup
John Coogan@johncoogan
“Yeah i was actually into slop back in 2015. Deep dream, pretty underground stuff. probably haven’t heard of it.”
English

@GregKamradt Man I’ve been sitting on this idea for so long. Glad you executed.
English

cards loaded
ui polish up next - going to mimic tbpn website style

Greg Kamradt@GregKamradt
ok pulling the trigger on this announcing so I'm publicly accountable it'll be my new nights/weekends garden project who knows a high quality custom card printer?
English












