I also set all delegation & subagents to DeepSeek v4 Flash so that my background work never burns OpenAI credits. I'm still fine-tuning the setup and considering LiteLLM to round out the routing layer. A multi-model dispatch that is easy on your wallet is the move ๐ฅ
Wired up Hermes with a multi-model stack:
Still a WIP, but DeepSeek v4 Pro via OpenRouter for primary, ChatGPT/Codex for heavy lifts, Cloudflare Workers AI for background compression. The cheap models trim context so the primary model stays focused and saves a ton of tokens.
@traversymedia Weโre basically back in 90โs, trading spacer GIFs & nested โ table tags for people creating giant prompts praying it one shots an entire enterprise app. Everyoneโs out here vibing their projects into the same spaghetti code everyone spent twenty years trying to clean up ๐
Very insightful advice.
Iโm running a two-tiered approach.
One folder is for frontier models that I want to archive (just in case). The second folder is for compressed (GGUF) models that I can actually use now through Ollama/LM Studio/llama.cpp.
Covering all the bases!
An important project I think everyone should do:
I got a 16TB hard drive and I'm downloading as many open source models as I can
This is insurance for the future. In case they are banned
If you have them already downloaded, they can't be taken away from you
And even if they aren't banned, it will be incredible to look back in 30 years and load up Qwen 3 just to see what life was like back then
If you have a smaller hard drive, I still think this is worth doing to stay independent and free, and having intelligence whenever need it. Download any models you can get your hands on from HuggingFace
It's preserving history and preserving your sovereignty
This is the first release, but it's been working great so far.
I would love to get some feedback from the Astro community. Feel free to open a #GitHub issue, or reply here. Any feedback is much appreciated!
#AIAgents#opensource#buildinpublic@astrodotbuild
I kept rebuilding the same SEO setup on all of my Astro sites. So I turned it into a repo! ๐๐ซก
Astro SEO Essentials is one command, fill in a few blanks, every page gets titles, social cards, sitemap, robots.txt & JSON-LD.
https://t.co/pnX9XbVuDi
#astrojs#webdev#SEO
cont.
I also added a Skill.md for AI agents.
Once its setup: Type /astro-seo-essentials in Claude Code (or $astro-seo-essentials in Codex) and the agent runs the exact same setup, making it an easy & repeatable process.
Hermes support is documented, and currently being tested.
@YS My biggest argument would be pricing... I can spin up multiple Wordpress sites and test an idea or product for practically free. $212 a mo is pretty hefty for the basic integrations on an e-com site.
@andrewpprice Ahhh yes! The Pinterest rabbit hole ๐ Thanks man! Iโm learning C4D now but now Blender is on my radar for rendering scenes and buildings because of your tutorials ๐๐ผ
@andrewpprice I just found your YouTube channel, great stuff! Iโm curious what you searched in Pinterest to find your inspiration for the โCreate a Subwayโ tutorial??