Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-08-28 · source-backed
PR 27342 adds grouped dynamic depthwise convolution and a candidate selector for DFlash2 draft models, reporting 1.81x on Apple M5 Pro running Qwen3.8-27B Q4_K_M at about 5.03 token acceptance, 1.77x to 1.85x across BF16 and Q8_0, and one Nvidia user seeing a consistent 2x decode speedup at every depth measured, holding at 32k context (GitHub). It auto-enables for DFlash2 checkpoints, so you get it without touching your launch command.
Each link below shares sources, entities, or timing with this story.
NVIDIA partners with Microsoft / Shared entities / Same source domain / Earlier coverage
Linked by a graph relationship (NVIDIA partners with Microsoft); both cover GitHub, Nvidia; reported by the same outlet (github.com).
NVIDIA released Nemotron / Shared entities / Same source domain / Earlier coverage
Linked by a graph relationship (NVIDIA released Nemotron); both cover GitHub, NVIDIA; reported by the same outlet (github.com).
NVIDIA uses Claude Code / Shared entities / Same source domain / Earlier coverage
Linked by a graph relationship (NVIDIA uses Claude Code); both cover GitHub, NVIDIA; reported by the same outlet (github.com).
Linked by a graph relationship (NVIDIA uses Claude Code); both cover GitHub, NVIDIA; reported by the same outlet (github.com).
Linked by a graph relationship (NVIDIA uses Claude Code); both cover GitHub, Qwen3; reported by the same outlet (github.com).
NVIDIA released DGX Spark / Shared entities / Same source domain / Earlier coverage
Linked by a graph relationship (NVIDIA released DGX Spark); both cover NVIDIA, Qwen3; reported by the same outlet (github.com).
NVIDIA released Nemotron / Shared entities / Same source domain / Earlier coverage
Linked by a graph relationship (NVIDIA released Nemotron); both cover GitHub, NVIDIA; reported by the same outlet (github.com).
NVIDIA invested in OpenAI / Shared entities / Earlier coverage / Tension
Linked by a graph relationship (NVIDIA invested in OpenAI); both cover GitHub, NVIDIA; earlier GitHub coverage from 2026-06-27.