Developing story · 2 episodes · 19 Oct to 19 Oct
Qwen Flash Next Hardware Optimization Race
The saga tracks the rapid ecosystem development surrounding the Qwen 3.8-Flash-Next model, beginning with the release of a specialized draft model by Inco AI. It currently stands with Ijah having launched a Metal-based inference engine to optimize the model's performance on Apple M5 hardware.
-
Ijah launches Metal‑based inference engine for Qwen 3.8‑Flash‑Next on Apple M5
The open‑source ijah engine runs on Apple M5 GPUs using raw Metal code and serves the Qwen3.8‑Flash‑Next model and its Swift‑1.5 fine‑tune via an OpenAI‑compatible HTTP API. It requires an Apple M5…
1 source primary source -
Inco AI releases DFlash 2 draft model for Qwen3.8-Flash-Next
DFlash 2 draft model is released as a repository that runs inside a speculative decoding server and drafts tokens for the target model Qwen3.8-Flash-Next. It predicts a whole block of tokens in a…
1 source primary source
Who and what