very excited to announce that @Mho_23 & i have cofounded VidLab...
the cheapest API to use any model with, even cheaper than the source
on top of that, we have tons of optimized systems for any type of creative that your agent can call via. the API
we are also giving away 1M free generations of our own model, VidLab 1.5, you can generate right away by signing up to https://t.co/At7H6GjmR0
NEW: Inkling 2 Mini is the most efficient model we've tested:
- #1 on efficiency (cost vs. intelligence)
- #9 on intelligence (78.4) overall
- #1 on intelligence among open-weights models
- Lowest cost per correct answer ($25.80)
- Biggest accuracy gain from adding web search
It uses a looped transformer, which runs its middle layers three times for every token. That lets a 26B-parameter model reason more without generating more tokens, which keeps costs low. A model that size can't store as many facts, so it relies on search to fill the gaps, and that's where it gains the most.
so here’s which one should you use according to use case
if the job is:
1 confusing:- use 1st (keep a human close)
2. clear & repeats a lot:- give more freedom to ai
that’s all
don’t make every ai fully free, & don’t control every single step either
just pick the right ai.
4. proactive (event driven)
this one doesn’t wait for you or an alarm
something happens (like someone sends) and the ai jumps into action
when it’s done, another ai checks the work
you only get called it something looks weird
this is the most independent ai all all