Qwen3.8-Flash-Next on any consumer hardware: one-click install for Windows / Linux. Strata inference engine, OpenAI/Anthropic API on localhost, optional image input.
C++
docs: the routing default is +67% with two conversations in flight
Niko1221f9907b6d9ce5
diff --git a/docs/release-notes-0.2.7.md b/docs/release-notes-0.2.7.mdindex 9f6ebd441..76289d720 100644--- a/docs/release-notes-0.2.7.md+++ b/docs/release-notes-0.2.7.md@@ -54,7 +54,8 @@ on the RTX 3060 12 GB (IQ3_XXS), five interleaved pairs of fresh engines:| prompt 4K, in a 4K context | 248.5 | **291.0 (+17.1%)** || decode behind a 4K document | 45.1 | **62.2 (+37.9%)** |-It holds where it matters most: on content the tuning was never fitted to it measures **+29.0%** (43.5 -> 55.0 tok/s on a+It holds where it matters most: with **two conversations in flight** it measures **+67% aggregate** (59.7 against 35.6 tok/s),+and on content the tuning was never fitted to it **+29.0%** (43.5 -> 55.0 tok/s on adocument prompt), and the whole recipe - routing, a workload-fitted expert profile, `--adapt-every 0`, `--vram-reserve-mib896` and the English draft vocabulary - gives **+26.5%** there and **+80%** on code in a sustained 80-request session.