Uploaded September 2026 | Updated November 2026, 4 days ago
Before joining Nebius, the Eigen AI team held the top spots on Artificial Analysis inference benchmarks across multiple models.
At Ray Summit 2026, Ryan Wang, SVP of Applied AI at Nebius, covers the inference framework techniques behind those results, from kernel and runtime optimization to NVFP4 quantization on Blackwell and speculative decoding, and what changes when you go from benchmark wins to production traffic at scale on Nebius Token Factory.
You'll leave with concrete optimization techniques and a clear sense of what production traffic actually demands.
Liked this video? Check out other Ray Summit breakout session recordings: youtube.com/playlist?list=PLZNaJyYZRPdo
Subscribe to our YouTube channel to stay up-to-date on the future of AI: youtube.com/c/anyscale
đź”— Connect with us:
LinkedIn: linkedin.com/company/joinanyscale
X: https://x.com/anyscalecompute
Website: anyscale.com
Before joining Nebius, the Eigen AI team held the top spots on Artificial Analysis inference benchmarks across multiple models.
At Ray Summit 2026, Ryan Wang, SVP of Applied AI at Nebius, covers the inference framework techniques behind those results, from kernel and runtime optimization to NVFP4 quantization on Blackwell and speculative decoding, and what changes when you go from benchmark wins to production traffic at scale on Nebius Token Factory.
You'll leave with concrete optimization techniques and a clear sense of what production traffic actually demands.
Liked this video? Check out other Ray Summit breakout session recordings: youtube.com/playlist?list=PLZNaJyYZRPdo
Subscribe to our YouTube channel to stay up-to-date on the future of AI: youtube.com/c/anyscale
đź”— Connect with us:
LinkedIn: linkedin.com/company/joinanyscale
X: https://x.com/anyscalecompute
Website: anyscale.com










