Skip to content

PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度

Support new model Qwen/Qwen3.6-35B-A3B (#1705)

Liang Lv2dda035b275a
diff --git a/auto_round/inference/convert_model.py b/auto_round/inference/convert_model.py
index e8a2779ec..4ae6405f0 100644
--- a/auto_round/inference/convert_model.py
+++ b/auto_round/inference/convert_model.py
@@ -179,6 +179,43 @@ def get_available_devices():
return devices
+def _remap_paths_for_text_model(model, quant_block_list, extra_config):
+ """Remap quantization paths when a composite model checkpoint is loaded as its text sub-model.
+
+ Uses Transformers' conversion_mapping WeightRenaming rules (e.g. "model.language_model" -> "model")
+ to fix path mismatches. Returns (remapped_block_list, remapped_extra_config).
+ """
+ try:
+ from transformers.conversion_mapping import get_checkpoint_conversion_mapping
+ except ImportError:
+ return quant_block_list, extra_config
+
+ model_type = getattr(getattr(model, "config", None), "model_type", None)
+ if model_type is None:
+ return quant_block_list, extra_config
+
+ mapping = get_checkpoint_conversion_mapping(model_type)
+ renamings = [r for r in mapping if type(r).__name__ == "WeightRenaming"]
+ if not renamings:
+ return quant_block_list, extra_config
+
+ def remap(path):
+ for r in renamings:
+ for src, tgt in zip(r.source_patterns, r.target_patterns):
+ new_path = re.sub(src, tgt, path)
+ if new_path != path:
+ return new_path
+ return path
+
+ new_block_list = [remap(b) for b in quant_block_list]
+ new_extra_config = {remap(k): v for k, v in extra_config.items()} if extra_config else extra_config
+
+ if new_block_list != quant_block_list:
+ logger.info(f"Remapped block_name_to_quantize: {quant_block_list} -> {new_block_list}")
+
+ return new_block_list, new_extra_config
+
+
def get_layer_config(model, quantization_config):
"""
get a layer-wise quantization configuration for a given model.
@@ -259,6 +296,17 @@ def get_layer_config(model, quantization_config):
# Load extra configuration if available
extra_config = getattr(quantization_config, "extra_config", {})
+ # When a composite model (e.g. VLM) is loaded as its text sub-model via AutoModelForCausalLM,
+ # block_name_to_quantize may still reference composite-level paths (e.g. "model.language_model.layers")
+ # while the actual module paths are "model.layers". Use conversion_mapping to remap if no layers matched.
+ if not layer_names and quant_block_list:
+ quant_block_list, extra_config = _remap_paths_for_text_model(model, quant_block_list, extra_config)
+ for n, m in model.named_modules():
+ if type(m) not in SUPPORTED_LAYER_TYPES:
+ continue
+ if check_start_with_block_name(n, quant_block_list):
+ layer_names.append(n)
+
# Process GPTQ format: identify modules that should be quantized
if getattr(quantization_config, "modules_in_block_to_quantize", None):
modules_in_block_to_quantize = flatten_list(