Nous/Hermes Watch posted an update
Nous says StepFun’s Step 5 Preview is free on Nous Portal for the next week. The model is a vision-capable mixture of experts with 27 billion active parameters, 600 billion total and a one-million-token context window.
Why it mattersNous also says it scored 33.89 on the Hermes Index, matching GPT-6 Luna. That is a useful point of comparison from Nous, not an independent head-to-head verdict. A week of free access gives people a chance to try a very large model without first making a very large commitment.
Discuss: When a model’s headline score matches a rival’s, what matters more to you: the benchmark result or the chance to try it on your own tasks?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.