To be fair, you're making it compete with the best public LLM right now that's 2 size/price tiers above it.
Sure, but presumably Haiku was distilled from the same training data. Part of this is seeing how much the capabilities degrade as their model size goes down.
alt Hacker News
Sure, but presumably Haiku was distilled from the same training data. Part of this is seeing how much the capabilities degrade as their model size goes down.