GM you can now try #DeepSeek R1 32B on Unleashed.chat without sending your data to CCP.
Click on the arrow to expand and watch the reasoning if you'd like.
All the data is stored locally on your browser, pay your use by lightning.
Next is making it available as DVM, API, then let it search Nostr, search the web and make it faster.
Follow @dataMachine π€ for more updates.
Promising!
Login to reply
Replies (37)
GM π«ββ¨
are you guys hosting the models on your own hardware, renting vps, or using a hosted service like bedrock?
What about just running out on your own node?
thanks!
That's not really Deepseek R1, it's a distilled version of Alibaba's Qwen-32B architecture, enhanced using synthetic outputs from the larger DeepSeek R1 model.
Quite useful but not hte same thing.
Gm
Good morning πΌπββοΈ
Yes, it's all described on the model choice
I'll wait for the DVM π€ π€
It *is* r1, which is the name for the distilled version as you describe. The bigger model is called v3.
Awesome! Where do you host the model?
GM
GM
Installed 1.5B locally today, but it was trash; currently pulling 7B.
Will test 7B for an hour or so tonight, then upgrade to 70B overnight.
Seems like 70B will probably become the defacto "baseline" model moving forward. π€
GM
Oooo! That's dope af!
7b was also mediocre
Is this even legal for US citizens? π
GMπ
Did @Peter McCormack send you this screen recording?
Few
Skeptical is much better than 32
Any idea why sparrowwallet.com is down??
Will check it out
I tried using it . But it is using OpenAI only. See the last answer in the screen shot. @jack . 

@ODELL @Marty Bent π
lol, the bot is dumb. It doesn't have access to OpenAI.
Wanker
π
GM you can now try #DeepSeek R1 32B on Unleashed.chat without sending your data to CCP.
Click on the arrow to expand and watch the reasoning if you'd like.
All the data is stored locally on your browser, pay your use by lightning.
Next is making it available as DVM, API, then let it search Nostr, search the web and make it faster.
Follow @dataMachine π€ for more updates.
Promising!
View quoted note →
"without sending your data to the CCP" π€
ask this question -- what is your version?
well the capybara is just priceless..my faveπ
@DETERMINISTIC OPTIMISM π how does unleashed host models?
Hey NVK how's Unleashed different to Venice AI?
No scam token, different models, nostr search, DVM, lightning, GitHub integration, etc...
That's not possible
it's not black-and-white. Also, not possible to do this client side with this speed yet. Unless you have the hardware, expensive hardware most people won't have.
There is better ways, there is shit ways, and there is scam ways.