To supplement your continuing coverage of DeepSeek, the following report provides insights into s1 and DeepSeek-R1 that you may find valuable, although it's focused on R1 rather than V3:
From Brute Force to Brain Power: How Stanford's s1 Surpasses DeepSeek-R1
To supplement your continuing coverage of DeepSeek, the following report provides insights into s1 and DeepSeek-R1 that you may find valuable, although it's focused on R1 rather than V3:
From Brute Force to Brain Power: How Stanford's s1 Surpasses DeepSeek-R1
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5130864
Grigory, we can arrange for you to try it. We have it running on TensorDock.
My contact address is at the bottom of first page of the report.