From 944408e93b1cda1266fda6c2f1aeca9b4c30ee75 Mon Sep 17 00:00:00 2001
From: Vladimir Mandic <mandic00@live.com>
Date: Tue, 10 Dec 2024 10:39:13 -0500
Subject: [PATCH] warn on quanto with offload

Signed-off-by: Vladimir Mandic <mandic00@live.com>
---
 modules/model_quant.py | 7 +++++--
 modules/shared.py      | 4 ++--
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/modules/model_quant.py b/modules/model_quant.py
index 9482fe898..03043b33a 100644
--- a/modules/model_quant.py
+++ b/modules/model_quant.py
@@ -58,11 +58,11 @@ def load_torchao(msg='', silent=False):
         import torchao
         ao = torchao
         fn = f'{sys._getframe(2).f_code.co_name}:{sys._getframe(1).f_code.co_name}' # pylint: disable=protected-access
-        log.debug(f'Quantization: type=quanto version={ao.__version__} fn={fn}') # pylint: disable=protected-access
+        log.debug(f'Quantization: type=torchao version={ao.__version__} fn={fn}') # pylint: disable=protected-access
         return ao
     except Exception as e:
         if len(msg) > 0:
-            log.error(f"{msg} failed to import optimum.quanto: {e}")
+            log.error(f"{msg} failed to import torchao: {e}")
         ao = None
         if not silent:
             raise
@@ -92,6 +92,7 @@ def load_bnb(msg='', silent=False):
 
 
 def load_quanto(msg='', silent=False):
+    from modules import shared
     global quanto # pylint: disable=global-statement
     if quanto is not None:
         return quanto
@@ -101,6 +102,8 @@ def load_quanto(msg='', silent=False):
         quanto = optimum_quanto
         fn = f'{sys._getframe(2).f_code.co_name}:{sys._getframe(1).f_code.co_name}' # pylint: disable=protected-access
         log.debug(f'Quantization: type=quanto version={quanto.__version__} fn={fn}') # pylint: disable=protected-access
+        if shared.opts.diffusers_offload_mode != 'none':
+            shared.log.error(f'Quantization: type=quanto offload={shared.opts.diffusers_offload_mode} not supported')
         return quanto
     except Exception as e:
         if len(msg) > 0:
diff --git a/modules/shared.py b/modules/shared.py
index 3d7571029..17db4595f 100644
--- a/modules/shared.py
+++ b/modules/shared.py
@@ -549,7 +549,7 @@ options_templates.update(options_section(('backends', "Backend Settings"), {
     "diffusers_fuse_projections": OptionInfo(False, "Fused projections"),
     "torch_expandable_segments": OptionInfo(False, "Expandable segments"),
     "cuda_mem_fraction": OptionInfo(0.0, "Memory limit", gr.Slider, {"minimum": 0, "maximum": 2.0, "step": 0.05}),
-    "torch_gc_threshold": OptionInfo(80, "GC threshold", gr.Slider, {"minimum": 0, "maximum": 100, "step": 1}),
+    "torch_gc_threshold": OptionInfo(70, "GC threshold", gr.Slider, {"minimum": 0, "maximum": 100, "step": 1}),
     "inference_mode": OptionInfo("no-grad", "Inference mode", gr.Radio, {"choices": ["no-grad", "inference-mode", "none"]}),
     "torch_malloc": OptionInfo("native", "Memory allocator", gr.Radio, {"choices": ['native', 'cudaMallocAsync'] }),
 
@@ -566,7 +566,7 @@ options_templates.update(options_section(('backends', "Backend Settings"), {
     "olive_cache_optimized": OptionInfo(True, 'Olive cache optimized models'),
 
     "ipex_sep": OptionInfo("<h2>IPEX</h2>", "", gr.HTML, {"visible": devices.backend == "ipex"}),
-    "ipex_optimize": OptionInfo([], "IPEX Optimize for Intel GPUs", gr.CheckboxGroup, {"choices": ["Model", "VAE", "Text Encoder", "Upscaler"], "visible": devices.backend == "ipex"}),
+    "ipex_optimize": OptionInfo([], "IPEX Optimize", gr.CheckboxGroup, {"choices": ["Model", "VAE", "Text Encoder", "Upscaler"], "visible": devices.backend == "ipex"}),
 
     "openvino_sep": OptionInfo("<h2>OpenVINO</h2>", "", gr.HTML, {"visible": cmd_opts.use_openvino}),
     "openvino_devices": OptionInfo([], "OpenVINO devices to use", gr.CheckboxGroup, {"choices": get_openvino_device_list() if cmd_opts.use_openvino else [], "visible": cmd_opts.use_openvino}), # pylint: disable=E0606