feat: add Apple Silicon DFlash benchmark planner (refs #152)

2026-04-21 22:00:22 -04:00
parent 492c1cdcfd
commit 636d294896
7 changed files with 385 additions and 6 deletions
--- a/evolution/quant_selector.py
+++ b/evolution/quant_selector.py
@@ -379,8 +379,8 @@ def select_quant_level(
            break

    if chosen is None:
-        # Nothing fits — pick the most aggressive compression
-        chosen = QUANT_LEVELS[-1]
+        # Nothing fits — pick the most aggressive compression, not the q4_0 fallback.
+        chosen = max(QUANT_LEVELS, key=lambda level: level.compression_ratio)
        logger.warning(f"No quant level fits in {memory_pool_gb:.1f}GB. Using {chosen.name}.")

    # Calculate final numbers