Phi-3-vision-128k-instruct-W4A16-G128

41
1
1 language
license:apache-2.0
by
RedHatAI
Language Model
OTHER
New
41 downloads
Early-stage
Try on Hugging FaceAdd to Compare
Edge AI:
Mobile
Laptop
Server
Unknown
Mobile
Laptop
Server
Quick Summary

Model Overview - Model Architecture: Phi-3-vision-128k-instruct - Input: Vision-Text - Output: Text - Model Optimizations: - Weight quantization: INT4 - Activat...

Training Data Analysis

🟡 Average (5.2/10)

Researched training datasets used by Phi-3-vision-128k-instruct-W4A16-G128 with quality assessment

Specialized For

code
general
science
multilingual

Training Datasets (3)

the pile
🟢 8/10
code
general
science
multilingual
Key Strengths
  • •Deliberate Diversity: Explicitly curated to include diverse content types (academia, code, Q&A, book...
  • •Documented Quality: Each component dataset is thoroughly documented with rationale for inclusion, en...
  • •Epoch Weighting: Component datasets receive different training epochs based on perceived quality, al...
common crawl
🔴 2.5/10
general
science
Key Strengths
  • •Scale and Accessibility: At 9.5+ petabytes, Common Crawl provides unprecedented scale for training d...
  • •Diversity: The dataset captures billions of web pages across multiple domains and content types, ena...
  • •Comprehensive Coverage: Despite limitations, Common Crawl attempts to represent the broader web acro...
Considerations
  • •Biased Coverage: The crawling process prioritizes frequently linked domains, making content from dig...
  • •Large-Scale Problematic Content: Contains significant amounts of hate speech, pornography, violent c...
wikipedia
🟡 5/10
science
multilingual
Key Strengths
  • •High-Quality Content: Wikipedia articles are subject to community review, fact-checking, and citatio...
  • •Multilingual Coverage: Available in 300+ languages, enabling training of models that understand and ...
  • •Structured Knowledge: Articles follow consistent formatting with clear sections, allowing models to ...
Considerations
  • •Language Inequality: Low-resource language editions have significantly lower quality, fewer articles...
  • •Biased Coverage: Reflects biases in contributor demographics; topics related to Western culture and ...

Explore our comprehensive training dataset analysis

View All Datasets

Code Examples

Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)
Creationpythontransformers
import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoProcessor

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot

# Load model.
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
    _attn_implementation="eager",
)
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
processor.chat_template = processor.tokenizer.chat_template

# Calibration dataset arguments
DATASET_ID = "lmms-lab/flickr30k"
DATASET_SPLIT = "test[:512]"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))

# Apply chat template and tokenize inputs.
def preprocess_and_tokenize(example):
    messages = [{"role": "user", "content": "<|image_1|>\nWhat does the image show?"}]
    text = processor.apply_chat_template(
        messages,
        add_generation_prompt=True,
    )
    images = example["image"]

    return processor(
        text=text,
        images=images,
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
    )

ds = ds.map(preprocess_and_tokenize, writer_batch_size=1, remove_columns=ds.column_names)

# Define a oneshot data collator for multimodal inputs.
def data_collator(batch):
    assert len(batch) == 1
    return {key: torch.tensor(value) for key, value in batch[0].items()}


# Recipe
recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Phi3DecoderLayer"],
    ignore=["lm_head", "re:model.vision_embed_tokens.*"],
)

# Perform oneshot
SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"

oneshot(
    model=model,
    processor=processor,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    trust_remote_code_model=True,
    data_collator=data_collator,
    output_dir=SAVE_DIR
)

Deploy This Model

Production-ready deployment in minutes

Together.ai

Instant API access to this model

Fastest API

Production-ready inference API. Start free, scale to millions.

Try Free API

Replicate

One-click model deployment

Easiest Setup

Run models in the cloud with simple API. No DevOps required.

Deploy Now

Disclosure: We may earn a commission from these partners. This helps keep LLMYourWay free.