4 ms·
> These models are our best yet thanks to distillation from Llama 4 Behemoth, a 288 billion active parameter model with 16 experts that is our most powerful yet
by scosman 2y ago
> These models are our best yet thanks to distillation from Llama 4 Behemoth, a 288 billion active parameter model with 16 experts that is our most powerful yet and among the world’s smartest LLMs. Llama 4 Behemoth outperforms GPT-4.5, Claude Sonnet 3.7, and Gemini 2.0 Pro on several STEM benchmarks. Llama 4 Behemoth is still training, and we’re excited to share more details about it even while it’s still in flight.
- senko 2y agoWith 2T params (!!), it better outperform everything else.
- amarcheschi 2y agoGiven that the comparison doesn't include O3 or gemini pro 2.5, I'd say it doesn't. Looking both at the comparison table available for llama 4 behemoth and gemini pro 2.5 it seems like at least a few of the comparable items might be won by gemini https://blog.google/technology/google-deepmind/gemini-model-thinking-updates-march-2025/ https://blog.google/technology/google-deepmind/gemini-model-...
- wmf 2y agoWe don't know how many params GPT-4, Claude, and Gemini are using so it could be in the ballpark.