Based on 30 visible X reactions from 136 accounts.
Ask a question below.
Published answers will appear here.
🚀 This feels like OpenAI's biggest workflow upgrade yet, not just another model release. With Sol pushing the frontier, Terra balancing performance and cost, and Luna optimizing speed, the focus is shifting from benchmark wins to real productivity. The real question is whether multi-agent workflows can consistently save hours not just seconds. #OpenAI #GPT56 #AI #GenAI
@rasbt Terra should never be used. It's pointless. Price and performance make the winner Luna ultra or Sol high and above
@rasbt concrete model recommendations with a chart instead of vibes is exactly what this space needs more of
@rasbt From my testing only Sol on high somewhat works, the rest is crap and IDK why they release it
The analysis recommends Luna over intermediate tiers like Terra.
@rasbt What a goddamn mess
@rasbt Uff! Fable API cost 🙁
For agentic coding, one can say: - Unless you need Terra Ultra perf, it's always better to use a Luna model with higher effort setting (same or better performance but cheaper). - Forget everything below Sol High, use Luna with higher effort settings here - Forget Sol Extra High, use Terra Ultra here - The extra cost of Sol Ultra is probably not worth it over Max
Meanwhile, the open models keep on keeping pace. It is as if there's no wall to incremental improvement. https://x.com/michaeljburry/status/2075565375274529099
The charts don't lie! (I suspect) https://twitter.com/rasbt/status/2075573860796436626
It’s simple https://twitter.com/rasbt/status/2075573860796436626
Based on 30 visible X reactions from 136 accounts.
Ask a question below.
Published answers will appear here.
@rasbt Uff! Fable API cost 🙁