Skip to content

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++

Gyro-M text: 16-21% slower than Q2_0 everywhere (setup's model list, strata_mcp, GYRO.md intro), as README and MODELS.md say

Niko1221328a3f5bd2f1
diff --git a/docs/GYRO.md b/docs/GYRO.md
index 61dc13417..e5ec06b5a 100644
--- a/docs/GYRO.md
+++ b/docs/GYRO.md
@@ -18,7 +18,7 @@ About half of each file is the n-gram (PLE) table: it stays on the SSD and is re
the file on a fast local drive. The memory figure is what the weights take once loaded.
Not defaults, by measurement (our evaluation, 2026-10): against Q2_0 at the same GPU bytes Gyro-M halves the KL
-divergence (0.159 against 0.312 on 128 chunks) and is level with IQ3_XXS (0.157), but decodes 17-21% slower on a 12 GB
+divergence (0.159 against 0.312 on 128 chunks) and is level with IQ3_XXS (0.157), but decodes 16-21% slower on a 12 GB
card (RTX 3060, measured before engine 0.2.2: 16% since) and an R9700, and its reasoning set scored lower than Q2_0's (49/60 against 57/60). Gyro-S is 13% worse
in KL than IQ2_XS (0.268 against 0.236) with 24% fewer bytes. So: pick Gyro-S where nothing bigger fits, Gyro-M when
the 20% memory saving over IQ3_XXS matters. Setup and the MCP server mention them only where the other sizes do not
diff --git a/setup.py b/setup.py
index 94a2bd937..0d37d7d74 100644
--- a/setup.py
+++ b/setup.py
@@ -3459,7 +3459,7 @@ def tier_note(ram: float, vram: float = 0.0, gyro: bool = True) -> str:
if not gyro or pick != ("swift", "IQ2_XS"):
return ""
return ("this PC's RAM + VRAM is in the 40 GB tier: Swift 1.5's IQ2_XS is the pick; Gyro-M (35 GB) is the other choice, "
- "close to IQ3_XXS in quality but 17-21% slower at writing answers than Q2_0")
+ "close to IQ3_XXS in quality but 16-21% slower at writing answers than Q2_0")
def recommended_size(family: str, names: list, ram: float, vram: float = 0.0) -> str:
diff --git a/tools/strata_mcp.py b/tools/strata_mcp.py
index 825518ca9..916bbf037 100644
--- a/tools/strata_mcp.py
+++ b/tools/strata_mcp.py
@@ -1121,7 +1121,7 @@ class Strata:
total = f"{ram:.0f} GB of RAM + {tvram:.0f} GB of VRAM = {ram + tvram:.0f} GB"
if model == "IQ2_XS":
why = (f"{total}, the 40 GB tier: Swift 1.5's IQ2_XS (thinks for a shorter time)"
- + ("; Gyro-M (35 GB) is the other choice there, close to IQ3_XXS in quality but 17-21% slower at "
+ + ("; Gyro-M (35 GB) is the other choice there, close to IQ3_XXS in quality but 16-21% slower at "
"writing answers than Q2_0" if "Gyro-M" in models and best.get("vendor") in ("nvidia", "amd", "intel") else ""))
elif model == "IQ3_XXS":
why = f"{total}, the 48 GB tier: IQ3_XXS (better quality, slower)"