Tweet épinglé
Harness AI
28 posts
Harness AI
@tryharness
A browser based screen aware AI assistant that lives in a floating PiP window, powered by on-chain inference 0xD3E592E728AE3461BD97c7A6B359E1043dd83bA3
United States Inscrit le Şubat 2025
6 Abonnements218 Abonnés
Gm gm!
A few folks have been asking about the plan for the harness token, so wanted to share some thoughts.
At the moment fees generated are helping fund managed inference, im super grateful to everyone who’s supported so far to make this possible!
Moving forward, users will be able to stake their tokens for usage credits that refresh each month. Fine details still need to be ironed out but that is the game plan.
Right now the focus is on getting user feedback improving the product and discovering a core group of users that get value out of what harness has to offer
Would appreciate any thoughts and feedback below!
English
@AskSurplus is insanely good tech.
Managed inference through harness routes through surplus markets first to ensure our users are getting the absolute best rates for all inference needs.
Still private through Venice’s infrastructure, but at 60-90% of the cost.
Check us out at tryharness.ai
Join the community at discord.gg/2VtEvAxUeE
Surplus Intelligence@AskSurplus
Surplus Intelligence Week 3 240+ models. 7,200+ suppliers. 420K+ requests. 13B+ tokens routed. Week 2 was already a record. Week 3 broke it again. Inference now has a live orderbook. Here's what the data says👇🧵
English
Wen harness?
Right now! Harness V1 is now live for everyone who has signed up for the waitlist!
Theres more!
The @tryharness discord community is now open! I will be adding more people to the waitlist through discord, so if you are interested in trying it out come through!
Check out a demo of a couple new features since the last demo video, I hope you like them 😄
I will also be posting more regularly on this account now that the main build is in a good place so follow to stay up to date!
The discord is the place for regular product updates, future plans, and general banter among the community.
Super excited to get this out to the world, gHarness 🚀
tryharness.ai
discord.gg/Dk5WWgP6Hh
English
Have been heads down building like a mad man this week so excited to share what’s new with harness
English
@Trend_Ron Will have some updates on business plan and token plan this week.
Been chatting with some community members in the dms to get feedback and iterate on ideas which has been super helpful.
One man show at the moment so doing my best to get everything aligned !
English
Agents are only as good as the context you give them. So why not give them the best context you can?
English
@jp54362 Quite a bit, especially if you are picking the right tool for the job !
English
We’re building a browser based ai assistant that is screen aware and lives in a floating PiP window. No downloads no extensions!
Since it’s screen aware privacy is a big concern, so we are using embedding models, and VLM models that can run locally and privately in your browser via webllm using your own hardware for inference!
For semantic searching over what you’ve seen and requests that don’t require reasoning or tool calling, local models are quite powerful!
English
Local AI is starting to feel more practical
love their tagline, LM Studio’s new app lets you run local models, in your pocket
when Android and who else is building surfaces like this? I want to try all of them
LM Studio@lmstudio
Meet LM Studio's mobile app. Your local models, now in your pocket.
English