@haider1 Sometime in the future when AI has taken all our jobs we may lament how we eagerly cheered on stronger and more capable models. Little did we know.
@meghrajai Generally my experience too. At least Opus admits they're good findings. I would rather have two medium models check each other than one max model going solo.
@vavanessadev I use Sol-6.1 on a $20 plan for reviewing Opus code. It finds a ton of issues and is practically free. I find Sol perfectly good for coding technical apps, same plan.
@luckeyfaraday I'd say Sol - the Anthropic guns might be marginally better, but you just can't ignore the Sol price. After all, they're not 10x better.
@rubenssoto_ai@IndependentEco Sometimes the latest model is better than a previous model at some things, regardless of official standing. That goes for Opus vs Fabel too.
@IndependentEco Maybe they’re in a deadlock- each waiting for the other to show their cards and then drop something presumably better. Realistically they probably can’t alter the model too much in a few days but they might tweak price and compute parameters - anything to steal the spotlight.
@DanielZambrini Great, be interesting to see. Yes, I think that would be closer to real world use and in a sense more fair to the models ( I suspect Sol 6.1 will shine here)