DFlash2 speeds Beam decode up to 3.03× across benchmarks, vLLM reports
vllm_project · x · 2026-10-16
vLLM reports DFlash2 accelerating Reflection Beam's decode path: 1.72–3.03× decode-only speedups on GSM8K, MATH500, HumanEval, MBPP and MT-Bench at concurrency 1/16/32, with GSM8K hitting 3.03×/2.51×/2.75×. Measured in a disaggregated prefill/decode setup excluding prefill time. Reflection also reports better throughput with K=7 even though the drafter was trained with block size 16.
Related event: DFlash2 Speeds Up Beam Decoding by Up to 3.03x(2 posts)→
More from Infra
- Neocloud stocks sink on Gartner's "80% will fail" report; analyst calls it garbage — McDonaghMatthew · 2026-10-16
- JapanFold Claims One Tenstorrent Server Beats an Nvidia DGX B200 on Bio Models at a Third of the Price — DavidBennett__ · 2026-10-16
- vLLM ships Day-0 RL support for Beam in SkyRL, +62% LoRA decode throughput — vllm_project · 2026-10-16
- Inside Beam: 501B params, only 23B active per token via 8-of-256 expert routing — vllm_project · 2026-10-16
- CoreWeave founders cashed out $3.3 billion before the stock's meltdown — SumitGup · 2026-10-16
- The Verge spent a week with Nvidia's RTX Spark Arm laptops: here's what we learned — tomwarren · 2026-10-16