Youtube has some good moments, it gave me an interesting video suggestion, @thecodacus ran Qwen3.6 35b on an old 1060 and he got 17 tps. I guess I might start digging up some of my old hardware and see what it can do. That 1060 never dreamt in its entire life that it would prove to be so useful.
Here's the link: https://t.co/FhMxUBEX0t
I'm glad to be able to run a local assistant using Llama-cpp, Qwen3-coder-30B-A3B and Morefine RTX 4090M 16GB VRAM + 64GB RAM.
> llama-server -m models/Qwen3-Coder-30B-A3B-Instruct-Q8_0.gguf --n-gpu-layers 999 --n-cpu-moe 28 --no-mmap --webui-mcp-proxy
Thanks @thecodacus
⚡Build your own MCP Server using AI (educational)
I use @Trae_ai and the MCP template from @cole_medin to create my own MCP servers, which then can be used within @n8n_io workflows.
https://t.co/2RPE5h274r
Created this game demo over the weekend without writing a single line of code! Built via bolt (Claude Sonnet 3.7) and powered by R3f + three.js
Try it yourself - dodge enemies with real-time physics. Here's how I created the game + a cloneable character controller project
1/7
Wow, gemini flash 2.0 is killing it
one single prompt "make a react landing page" in @boltdotdiy , also check the setting to get this result in the reply
#aicoding#ai
@alexcovo_eth@deepseek_ai@boltdotdiy check out the latest version with "context optimization" and "auto select Template"
turned on.
also not it separates the think section and output sections
@deepseek_ai Deepseek-R1 has to be the politest and understanding model ever. I mean I wish I could offer it tea or food or something. I was able to make myself lunch, relax as it churned away at creating an app with @boltdotdiy. Even though I have no idea what I'm doing, it's gently holding my hand and guiding me. You're my best friend coding buddy now. 🥹
FU prick developers who use to give me an attitude when I asked a question. You know who you are. 😂😂😂
@NoBanksNearby@boltdotdiy Hi @NoBanksNearby,
can you try the official prebuilt docker image
and if if you are still getting this issue ? https://t.co/zrKk2nX8WY
@tvykruta@RnaudBertrand the intelligence of a model should not be tested by how accurately it can recall data from its training set.
but we should be testing the model with some critical thinking challenges which does not require it to recall accurate data, but ability to come up with a solution
It's only been just over 48 hours since DeepSeek dropped DeepSeek-R1, fully open source reasoning model.
And people are getting crazy creative with it.
10 wild examples:
here are some outputs of @boltdotdiy with #DeepSeekR1 with single prompt (added setting configuration for this to work)
thanks to @deepseek for open sourcing such a brilliant model and pushing the wall of "open ai" further
#boltdiy#bolt