update examples imports

kylesayrs · kylesayrs · commit 74aa7c92def6 · 2025-06-16T15:12:08.000-04:00
Signed-off-by: Kyle Sayers &lt;kylesayrs@gmail.com&gt;
diff --git a/examples/awq/qwen3_moe_example.py b/examples/awq/qwen3_moe_example.py
@@ -3,7 +3,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.awq import AWQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Select model and load it.
 MODEL_ID = "Qwen/Qwen3-30B-A3B"
diff --git a/examples/multimodal_audio/whisper_example.py b/examples/multimodal_audio/whisper_example.py
@@ -4,7 +4,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Select model and load it.
 MODEL_ID = "openai/whisper-large-v3"
diff --git a/examples/multimodal_vision/gemma3_example.py b/examples/multimodal_vision/gemma3_example.py
@@ -5,7 +5,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Load model.
 model_id = "google/gemma-3-4b-it"
diff --git a/examples/multimodal_vision/idefics3_example.py b/examples/multimodal_vision/idefics3_example.py
@@ -6,7 +6,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Load model.
 model_id = "HuggingFaceM4/Idefics3-8B-Llama3"  # or "HuggingFaceTB/SmolVLM-Instruct"
diff --git a/examples/multimodal_vision/llava_example.py b/examples/multimodal_vision/llava_example.py
@@ -5,7 +5,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Load model.
 model_id = "llava-hf/llava-1.5-7b-hf"
diff --git a/examples/multimodal_vision/mistral3_example.py b/examples/multimodal_vision/mistral3_example.py
@@ -8,7 +8,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Load model.
 model_id = "mistralai/Mistral-Small-3.1-24B-Instruct-2503"
diff --git a/examples/multimodal_vision/mllama_example.py b/examples/multimodal_vision/mllama_example.py
@@ -5,7 +5,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Load model.
 model_id = "meta-llama/Llama-3.2-11B-Vision-Instruct"
diff --git a/examples/multimodal_vision/phi3_vision_example.py b/examples/multimodal_vision/phi3_vision_example.py
@@ -7,7 +7,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Load model.
 model_id = "microsoft/Phi-3-vision-128k-instruct"
diff --git a/examples/multimodal_vision/pixtral_example.py b/examples/multimodal_vision/pixtral_example.py
@@ -5,7 +5,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Load model.
 model_id = "mgoin/pixtral-12b"
diff --git a/examples/multimodal_vision/qwen2_vl_example.py b/examples/multimodal_vision/qwen2_vl_example.py
@@ -8,7 +8,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Load model.
 model_id = "Qwen/Qwen2-VL-2B-Instruct"
diff --git a/examples/multimodal_vision/qwen_2_5_vl_example.py b/examples/multimodal_vision/qwen_2_5_vl_example.py
@@ -8,7 +8,7 @@
 
 from llmcompressor.modifiers.quantization import GPTQModifier
 from llmcompressor.transformers import oneshot
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Load model.
 model_id = "Qwen/Qwen2.5-VL-7B-Instruct"
diff --git a/examples/quantization_2of4_sparse_w4a16/llama7b_sparse_w4a16.py b/examples/quantization_2of4_sparse_w4a16/llama7b_sparse_w4a16.py
@@ -3,7 +3,7 @@
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
 from llmcompressor import oneshot, train
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # load the model in as bfloat16 to save on memory and compute
 model_stub = "neuralmagic/Llama-2-7b-ultrachat200k"
diff --git a/examples/quantization_kv_cache/gemma2_fp8_kv_example.py b/examples/quantization_kv_cache/gemma2_fp8_kv_example.py
@@ -2,7 +2,7 @@
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
 from llmcompressor import oneshot
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Select model and load it.
 MODEL_ID = "google/gemma-2-9b-it"
diff --git a/examples/quantization_kv_cache/llama3_fp8_kv_example.py b/examples/quantization_kv_cache/llama3_fp8_kv_example.py
@@ -3,7 +3,7 @@
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
 from llmcompressor import oneshot
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Select model and load it.
 MODEL_ID = "meta-llama/Meta-Llama-3-8B-Instruct"
diff --git a/examples/quantization_kv_cache/phi3.5_fp8_kv_example.py b/examples/quantization_kv_cache/phi3.5_fp8_kv_example.py
@@ -2,7 +2,7 @@
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
 from llmcompressor import oneshot
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Select model and load it.
 # Phi-3.5 is a special case for KV cache quantization because it has
diff --git a/examples/quantization_w4a16/llama3_example.py b/examples/quantization_w4a16/llama3_example.py
@@ -3,7 +3,7 @@
 
 from llmcompressor.modifiers.quantization import GPTQModifier
 from llmcompressor.transformers import oneshot
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Select model and load it.
 model_id = "meta-llama/Meta-Llama-3-8B-Instruct"
diff --git a/examples/quantization_w4a16_fp4/llama3_example.py b/examples/quantization_w4a16_fp4/llama3_example.py
@@ -2,7 +2,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 MODEL_ID = "meta-llama/Llama-3.1-8B-Instruct"
 
diff --git a/examples/quantization_w4a4_fp4/llama3_example.py b/examples/quantization_w4a4_fp4/llama3_example.py
@@ -3,7 +3,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 MODEL_ID = "meta-llama/Meta-Llama-3-8B-Instruct"
 
diff --git a/examples/quantization_w8a8_fp8/gemma2_example.py b/examples/quantization_w8a8_fp8/gemma2_example.py
@@ -2,7 +2,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 MODEL_ID = "google/gemma-2-27b-it"
 
diff --git a/examples/quantization_w8a8_fp8/llama3.2_vision_example.py b/examples/quantization_w8a8_fp8/llama3.2_vision_example.py
@@ -2,7 +2,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 MODEL_ID = "meta-llama/Llama-3.2-11B-Vision-Instruct"
 
diff --git a/examples/quantization_w8a8_fp8/llama3_example.py b/examples/quantization_w8a8_fp8/llama3_example.py
@@ -2,7 +2,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 MODEL_ID = "meta-llama/Meta-Llama-3-8B-Instruct"
 
diff --git a/examples/quantization_w8a8_fp8/llava1.5_example.py b/examples/quantization_w8a8_fp8/llava1.5_example.py
@@ -2,7 +2,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 MODEL_ID = "llava-hf/llava-1.5-7b-hf"
 
diff --git a/examples/quantization_w8a8_fp8/qwen2vl_example.py b/examples/quantization_w8a8_fp8/qwen2vl_example.py
@@ -2,7 +2,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 MODEL_ID = "Qwen/Qwen2-VL-7B-Instruct"
 
diff --git a/examples/quantization_w8a8_fp8/whisper_example.py b/examples/quantization_w8a8_fp8/whisper_example.py
@@ -3,7 +3,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 MODEL_ID = "openai/whisper-large-v2"
 
diff --git a/examples/quantization_w8a8_int8/gemma2_example.py b/examples/quantization_w8a8_int8/gemma2_example.py
@@ -3,7 +3,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # 1) Select model and load it.
 MODEL_ID = "google/gemma-2-2b-it"
diff --git a/examples/quantization_w8a8_int8/llama3_example.py b/examples/quantization_w8a8_int8/llama3_example.py
@@ -4,7 +4,7 @@
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
 from llmcompressor.modifiers.smoothquant import SmoothQuantModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Select model and load it.
 MODEL_ID = "meta-llama/Meta-Llama-3-8B-Instruct"
diff --git a/examples/quantizing_moe/deepseek_moe_w4a16.py b/examples/quantizing_moe/deepseek_moe_w4a16.py
@@ -4,7 +4,7 @@
 from transformers import AutoModelForCausalLM, AutoTokenizer, __version__
 
 from llmcompressor import oneshot
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # NOTE: transformers 4.49.0 has an attribute error with DeepSeek.
 # Please consider either downgrading your transformers version to a
diff --git a/examples/quantizing_moe/deepseek_moe_w8a8_fp8.py b/examples/quantizing_moe/deepseek_moe_w8a8_fp8.py
@@ -4,7 +4,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # NOTE: transformers 4.49.0 has an attribute error with DeepSeek.
 # Please consider either downgrading your transformers version to a
diff --git a/examples/quantizing_moe/deepseek_moe_w8a8_int8.py b/examples/quantizing_moe/deepseek_moe_w8a8_int8.py
@@ -5,7 +5,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import GPTQModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # NOTE: transformers 4.49.0 has an attribute error with DeepSeek.
 # Please consider either downgrading your transformers version to a
diff --git a/examples/quantizing_moe/mixtral_moe_w8a8_fp8.py b/examples/quantizing_moe/mixtral_moe_w8a8_fp8.py
@@ -5,7 +5,7 @@
 
 from llmcompressor import oneshot
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 MODEL_ID = "mistralai/Mixtral-8x7B-Instruct-v0.1"
 
diff --git a/examples/quantizing_moe/qwen_moe_w4a16.py b/examples/quantizing_moe/qwen_moe_w4a16.py
@@ -4,7 +4,7 @@
 
 from llmcompressor.modifiers.quantization import GPTQModifier
 from llmcompressor.transformers import oneshot
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # select a Mixture of Experts model for quantization
 MODEL_ID = "Qwen/Qwen1.5-MoE-A2.7B-Chat"
diff --git a/examples/sparse_2of4_quantization_fp8/llama3_8b_2of4.py b/examples/sparse_2of4_quantization_fp8/llama3_8b_2of4.py
@@ -6,7 +6,7 @@
 from llmcompressor import oneshot
 from llmcompressor.modifiers.obcq import SparseGPTModifier
 from llmcompressor.modifiers.quantization import QuantizationModifier
-from llmcompressor.utils.dev import dispatch_for_generation
+from llmcompressor.utils import dispatch_for_generation
 
 # Configuration
 MODEL_ID = "meta-llama/Meta-Llama-3-8B-Instruct"
diff --git a/src/llmcompressor/utils/__init__.py b/src/llmcompressor/utils/__init__.py
@@ -4,4 +4,5 @@
 
 # flake8: noqa
 
+from .dev import *
 from .helpers import *
