DeepSeek v4 flash doesn't make sense now after the GLM 5.3 flash(Ox Alpha) drop.
The speed of which a model can become irrelevant is insane Literally less than a month.
I am too broke to top up credits for API rate. Essentially relying on free models right now. They're not bad and sometimes decent models giveaway takes place: ox alpha an minimax right now are incredible workhorses.
I am too broke to top up credits for API rate. Essentially relying on free models right now. They're not bad and sometimes decent models giveaway takes place: ox alpha an minimax right now are incredible workhorses.
@joyofcodedev https://t.co/AoT6pTSIy7
The 3 bottom (hy3, Muse and Mimo) feels near unlimited + keep watching on the free models (ex: Ox Alpha is the hype righn now but there are many others free too). The rest, avoid ❌
Otherwise, the sub won't last an afternoon.
Muse Spark 1.2 Contributor is now available on OpenCode Go
this model will train on your data so it requires explicit opt-in and is restricted in some regions
in exchange you get incredible usage limits
@mattpocockuk I considered a VM/VPS but finally, nah...
What I did instead:
Dell optilex 3000 i5 12500T upgraded to 64GB of RAM running 24*7 at home. I can acces it with ssh wherever I want or through my Discord server
I tried 5USD of credits on @umans_ai , it ran out in less that 2 hours while only using deepseek v4 flash. Holly cow!! not for me I guess. I am continuing exploring other providers
I tried 5USD of credits on @umans_ai , it ran out in less that 2 hours while only using deepseek v4 flash. Holly cow!! not for me I guess. I am continuing exploring other providers