Reflection AI Beam: Open-Weight Launch Claims Checked
Reflection AI says its 501B Beam model matches GLM 5.2 on reasoning at a fraction of the compute. Here is what is verified and what is not.
5 articles on Model Drop tagged "mixture of experts."
5 articles
Reflection AI says its 501B Beam model matches GLM 5.2 on reasoning at a fraction of the compute. Here is what is verified and what is not.
Flint activates just 2.5 billion of its 40 billion parameters and ships under Apache 2.0. The speed and benchmark numbers are all self-reported, so here is what holds up.
Liquid AI's largest and last LFM2.5 model trades knowledge for speed. We read the model card, the pricing and the benchmark gaps so you know where it fits.
Mistral opened a hosted preview of its trillion-parameter Large 4. The weights, license, and independent benchmarks are still to come.
MoE gives you a small model's speed with an enormous model's memory footprint. That tradeoff decides more deployments than quality does.