Internvl3 9B AWQ

🚀 InternVL3-9B

InternVL3-9B 是一個先進的多模態大語言模型（MLLM），相比前代模型，它在多模態感知、推理等能力上有顯著提升，還拓展了工具使用、GUI 代理、工業圖像分析等多模態能力。此外，該模型在文本性能上也表現出色，優於 Qwen2.5 系列。

🚀 快速開始

我們提供了使用 transformers 運行 InternVL3-9B 的示例代碼。

⚠️ 重要提示

請使用 transformers>=4.37.2 以確保模型正常工作。

模型加載

16 位（bf16 / fp16）

import torch
from transformers import AutoTokenizer, AutoModel
path = "OpenGVLab/InternVL3-9B"
model = AutoModel.from_pretrained(
    path,
    torch_dtype=torch.bfloat16,
    low_cpu_mem_usage=True,
    use_flash_attn=True,
    trust_remote_code=True).eval().cuda()

BNB 8 位量化

import torch
from transformers import AutoTokenizer, AutoModel
path = "OpenGVLab/InternVL3-9B"
model = AutoModel.from_pretrained(
    path,
    torch_dtype=torch.bfloat16,
    load_in_8bit=True,
    low_cpu_mem_usage=True,
    use_flash_attn=True,
    trust_remote_code=True).eval()

多 GPU 情況

import math
import torch
from transformers import AutoTokenizer, AutoModel

def split_model(model_name):
    device_map = {}
    world_size = torch.cuda.device_count()
    config = AutoConfig.from_pretrained(model_path, trust_remote_code=True)
    num_layers = config.llm_config.num_hidden_layers
    # Since the first GPU will be used for ViT, treat it as half a GPU.
    num_layers_per_gpu = math.ceil(num_layers / (world_size - 0.5))
    num_layers_per_gpu = [num_layers_per_gpu] * world_size
    num_layers_per_gpu[0] = math.ceil(num_layers_per_gpu[0] * 0.5)
    layer_cnt = 0
    for i, num_layer in enumerate(num_layers_per_gpu):
        for j in range(num_layer):
            device_map[f'language_model.model.layers.{layer_cnt}'] = i
            layer_cnt += 1
    device_map['vision_model'] = 0
    device_map['mlp1'] = 0
    device_map['language_model.model.tok_embeddings'] = 0
    device_map['language_model.model.embed_tokens'] = 0
    device_map['language_model.output'] = 0
    device_map['language_model.model.norm'] = 0
    device_map['language_model.model.rotary_emb'] = 0
    device_map['language_model.lm_head'] = 0
    device_map[f'language_model.model.layers.{num_layers - 1}'] = 0

    return device_map

path = "OpenGVLab/InternVL3-9B"
device_map = split_model('InternVL3-9B')
model = AutoModel.from_pretrained(
    path,
    torch_dtype=torch.bfloat16,
    low_cpu_mem_usage=True,
    use_flash_attn=True,
    trust_remote_code=True,
    device_map=device_map).eval()

使用 Transformers 進行推理

import math
import numpy as np
import torch
import torchvision.transforms as T
from decord import VideoReader, cpu
from PIL import Image
from torchvision.transforms.functional import InterpolationMode
from transformers import AutoModel, AutoTokenizer

IMAGENET_MEAN = (0.485, 0.456, 0.406)
IMAGENET_STD = (0.229, 0.224, 0.225)

def build_transform(input_size):
    MEAN, STD = IMAGENET_MEAN, IMAGENET_STD
    transform = T.Compose([
        T.Lambda(lambda img: img.convert('RGB') if img.mode != 'RGB' else img),
        T.Resize((input_size, input_size), interpolation=InterpolationMode.BICUBIC),
        T.ToTensor(),
        T.Normalize(mean=MEAN, std=STD)
    ])
    return transform

def find_closest_aspect_ratio(aspect_ratio, target_ratios, width, height, image_size):
    best_ratio_diff = float('inf')
    best_ratio = (1, 1)
    area = width * height
    for ratio in target_ratios:
        target_aspect_ratio = ratio[0] / ratio[1]
        ratio_diff = abs(aspect_ratio - target_aspect_ratio)
        if ratio_diff < best_ratio_diff:
            best_ratio_diff = ratio_diff
            best_ratio = ratio
        elif ratio_diff == best_ratio_diff:
            if area > 0.5 * image_size * image_size * ratio[0] * ratio[1]:
                best_ratio = ratio
    return best_ratio

def dynamic_preprocess(image, min_num=1, max_num=12, image_size=448, use_thumbnail=False):
    orig_width, orig_height = image.size
    aspect_ratio = orig_width / orig_height

    # calculate the existing image aspect ratio
    target_ratios = set(
        (i, j) for n in range(min_num, max_num + 1) for i in range(1, n + 1) for j in range(1, n + 1) if
        i * j <= max_num and i * j >= min_num)
    target_ratios = sorted(target_ratios, key=lambda x: x[0] * x[1])

    # find the closest aspect ratio to the target
    target_aspect_ratio = find_closest_aspect_ratio(
        aspect_ratio, target_ratios, orig_width, orig_height, image_size)

    # calculate the target width and height
    target_width = image_size * target_aspect_ratio[0]
    target_height = image_size * target_aspect_ratio[1]
    blocks = target_aspect_ratio[0] * target_aspect_ratio[1]

    # resize the image
    resized_img = image.resize((target_width, target_height))
    processed_images = []
    for i in range(blocks):
        box = (
            (i % (target_width // image_size)) * image_size,
            (i // (target_width // image_size)) * image_size,
            ((i % (target_width // image_size)) + 1) * image_size,
            ((i // (target_width // image_size)) + 1) * image_size
        )
        # split the image
        split_img = resized_img.crop(box)
        processed_images.append(split_img)
    assert len(processed_images) == blocks
    if use_thumbnail and len(processed_images) != 1:
        thumbnail_img = image.resize((image_size, image_size))
        processed_images.append(thumbnail_img)
    return processed_images

def load_image(image_file, input_size=448, max_num=12):
    image = Image.open(image_file).convert('RGB')
    transform = build_transform(input_size=input_size)
    images = dynamic_preprocess(image, image_size=input_size, use_thumbnail=True, max_num=max_num)
    pixel_values = [transform(image) for image in images]
    pixel_values = torch.stack(pixel_values)
    return pixel_values

def split_model(model_name):
    device_map = {}
    world_size = torch.cuda.device_count()
    config = AutoConfig.from_pretrained(model_path, trust_remote_code=True)
    num_layers = config.llm_config.num_hidden_layers
    # Since the first GPU will be used for ViT, treat it as half a GPU.
    num_layers_per_gpu = math.ceil(num_layers / (world_size - 0.5))
    num_layers_per_gpu = [num_layers_per_gpu] * world_size
    num_layers_per_gpu[0] = math.ceil(num_layers_per_gpu[0] * 0.5)
    layer_cnt = 0
    for i, num_layer in enumerate(num_layers_per_gpu):
        for j in range(num_layer):
            device_map[f'language_model.model.layers.{layer_cnt}'] = i
            layer_cnt += 1
    device_map['vision_model'] = 0
    device_map['mlp1'] = 0
    device_map['language_model.model.tok_embeddings'] = 0
    device_map['language_model.model.embed_tokens'] = 0
    device_map['language_model.output'] = 0
    device_map['language_model.model.norm'] = 0
    device_map['language_model.model.rotary_emb'] = 0
    device_map['language_model.lm_head'] = 0
    device_map[f'language_model.model.layers.{num_layers - 1}'] = 0

    return device_map

# If you set `load_in_8bit=True`, you will need two 80GB GPUs.
# If you set `load_in_8bit=False`, you will need at least three 80GB GPUs.
path = 'OpenGVLab/InternVL3-9B'
device_map = split_model('InternVL3-9B')
model = AutoModel.from_pretrained(
    path,
    torch_dtype=torch.bfloat16,
    load_in_8bit=False,
    low_cpu_mem_usage=True,
    use_flash_attn=True,
    trust_remote_code=True,
    device_map=device_map).eval()
tokenizer = AutoTokenizer.from_pretrained(path, trust_remote_code=True, use_fast=False)

# set the max number of tiles in `max_num`
pixel_values = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
generation_config = dict(max_new_tokens=1024, do_sample=True)

# pure-text conversation (純文本對話)
question = 'Hello, who are you?'
response, history = model.chat(tokenizer, None, question, generation_config, history=None, return_history=True)
print(f'User: {question}\nAssistant: {response}')

question = 'Can you tell me a story?'
response, history = model.chat(tokenizer, None, question, generation_config, history=history, return_history=True)
print(f'User: {question}\nAssistant: {response}')

# single-image single-round conversation (單圖單輪對話)
question = '<image>\nPlease describe the image shortly.'
response = model.chat(tokenizer, pixel_values, question, generation_config)
print(f'User: {question}\nAssistant: {response}')

# single-image multi-round conversation (單圖多輪對話)
question = '<image>\nPlease describe the image in detail.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config, history=None, return_history=True)
print(f'User: {question}\nAssistant: {response}')

question = 'Please write a poem according to the image.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config, history=history, return_history=True)
print(f'User: {question}\nAssistant: {response}')

# multi-image multi-round conversation, combined images (多圖多輪對話，拼接圖像)
pixel_values1 = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
pixel_values2 = load_image('./examples/image2.jpg', max_num=12).to(torch.bfloat16).cuda()
pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)

question = '<image>\nDescribe the two images in detail.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               history=None, return_history=True)
print(f'User: {question}\nAssistant: {response}')

question = 'What are the similarities and differences between these two images.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               history=history, return_history=True)
print(f'User: {question}\nAssistant: {response}')

# multi-image multi-round conversation, separate images (多圖多輪對話，獨立圖像)
pixel_values1 = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
pixel_values2 = load_image('./examples/image2.jpg', max_num=12).to(torch.bfloat16).cuda()
pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)
num_patches_list = [pixel_values1.size(0), pixel_values2.size(0)]

question = 'Image-1: <image>\nImage-2: <image>\nDescribe the two images in detail.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               num_patches_list=num_patches_list,
                               history=None, return_history=True)
print(f'User: {question}\nAssistant: {response}')

question = 'What are the similarities and differences between these two images.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               num_patches_list=num_patches_list,
                               history=history, return_history=True)
print(f'User: {question}\nAssistant: {response}')

# batch inference, single image per sample (單圖批處理)
pixel_values1 = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
pixel_values2 = load_image('./examples/image2.jpg', max_num=12).to(torch.bfloat16).cuda()
num_patches_list = [pixel_values1.size(0), pixel_values2.size(0)]
pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)

questions = ['<image>\nDescribe the image in detail.'] * len(num_patches_list)
responses = model.batch_chat(tokenizer, pixel_values,
                             num_patches_list=num_patches_list,
                             questions=questions,
                             generation_config=generation_config)
for question, response in zip(questions, responses):
    print(f'User: {question}\nAssistant: {response}')

# video multi-round conversation (視頻多輪對話)
def get_index(bound, fps, max_frame, first_idx=0, num_segments=32):
    if bound:
        start, end = bound[0], bound[1]
    else:
        start, end = -100000, 100000
    start_idx = max(first_idx, round(start * fps))
    end_idx = min(round(end * fps), max_frame)
    seg_size = float(end_idx - start_idx) / num_segments
    frame_indices = np.array([
        int(start_idx + (seg_size / 2) + np.round(seg_size * idx))
        for idx in range(num_segments)
    ])
    return frame_indices

def load_video(video_path, bound=None, input_size=448, max_num=1, num_segments=32):
    vr = VideoReader(video_path, ctx=cpu(0), num_threads=1)
    max_frame = len(vr) - 1
    fps = float(vr.get_avg_fps())

    pixel_values_list, num_patches_list = [], []
    transform = build_transform(input_size=input_size)
    frame_indices = get_index(bound, fps, max_frame, first_idx=0, num_segments=num_segments)
    for frame_index in frame_indices:
        img = Image.fromarray(vr[frame_index].asnumpy()).convert('RGB')
        img = dynamic_preprocess(img, image_size=input_size, use_thumbnail=True, max_num=max_num)
        pixel_values = [transform(tile) for tile in img]
        pixel_values = torch.stack(pixel_values)
        num_patches_list.append(pixel_values.shape[0])
        pixel_values_list.append(pixel_values)
    pixel_values = torch.cat(pixel_values_list)
    return pixel_values, num_patches_list

video_path = './examples/red-panda.mp4'
pixel_values, num_patches_list = load_video(video_path, num_segments=8, max_num=1)
pixel_values = pixel_values.to(torch.bfloat16).cuda()
video_prefix = ''.join([f'Frame{i+1}: <image>\n' for i in range(len(num_patches_list))])
question = video_prefix + 'What is the red panda doing?'
# Frame1: <image>\nFrame2: <image>\n...\nFrame8: <image>\n{question}
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               num_patches_list=num_patches_list, history=None, return_history=True)
print(f'User: {question}\nAssistant: {response}')

question = 'Describe this video in detail.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               num_patches_list=num_patches_list, history=history, return_history=True)
print(f'User: {question}\nAssistant: {response}')

流式輸出

from transformers import TextIteratorStreamer
from threading import Thread

# Initialize the streamer
streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True, timeout=10)
# Define the generation configuration
generation_config = dict(max_new_tokens=1024, do_sample=False, streamer=streamer)
# Start the model chat in a separate thread
thread = Thread(target=model.chat, kwargs=dict(
    tokenizer=tokenizer, pixel_values=pixel_values, question=question,
    history=None, return_history=False, generation_config=generation_config,
))
thread.start()

# Initialize an empty string to store the generated text
generated_text = ''
# Loop through the streamer to get the new text as it is generated
for new_text in streamer:
    if new_text == model.conv_template.sep:
        break
    generated_text += new_text
    print(new_text, end='', flush=True)  # Print each new chunk of generated text on the same line

✨ 主要特性

• 卓越的多模態能力：相比 InternVL 2.5，InternVL3 展現出更出色的多模態感知和推理能力，還將多模態能力拓展到工具使用、GUI 代理、工業圖像分析、3D 視覺感知等領域。
• 優秀的文本性能：通過原生多模態預訓練，InternVL3 系列在整體文本性能上優於 Qwen2.5 系列。
• 靈活的模型架構：沿用 “ViT - MLP - LLM” 範式，集成新的增量預訓練 InternViT 和多種預訓練 LLM。
• 先進的訓練策略：採用原生多模態預訓練、監督微調、混合偏好優化和測試時縮放等策略，提升模型性能。

📚 詳細文檔

InternVL3 家族

以下表格概述了 InternVL3 系列：

模型名稱	視覺部分	語言部分	HF 鏈接
InternVL3 - 1B	[InternViT - 300M - 448px - V2_5](https://huggingface.co/OpenGVLab/InternViT - 300M - 448px - V2_5)	[Qwen2.5 - 0.5B](https://huggingface.co/Qwen/Qwen2.5 - 0.5B)	[🤗 link](https://huggingface.co/OpenGVLab/InternVL3 - 1B)
InternVL3 - 2B	[InternViT - 300M - 448px - V2_5](https://huggingface.co/OpenGVLab/InternViT - 300M - 448px - V2_5)	[Qwen2.5 - 1.5B](https://huggingface.co/Qwen/Qwen2.5 - 1.5B)	[🤗 link](https://huggingface.co/OpenGVLab/InternVL3 - 2B)
InternVL3 - 8B	[InternViT - 300M - 448px - V2_5](https://huggingface.co/OpenGVLab/InternViT - 300M - 448px - V2_5)	[Qwen2.5 - 7B](https://huggingface.co/Qwen/Qwen2.5 - 7B)	[🤗 link](https://huggingface.co/OpenGVLab/InternVL3 - 8B)
InternVL3 - 9B	[InternViT - 300M - 448px - V2_5](https://huggingface.co/OpenGVLab/InternViT - 300M - 448px - V2_5)	[internlm3 - 8b - instruct](https://huggingface.co/internlm/internlm3 - 8b - instruct)	[🤗 link](https://huggingface.co/OpenGVLab/InternVL3 - 9B)
InternVL3 - 14B	[InternViT - 300M - 448px - V2_5](https://huggingface.co/OpenGVLab/InternViT - 300M - 448px - V2_5)	[Qwen2.5 - 14B](https://huggingface.co/Qwen/Qwen2.5 - 14B)	[🤗 link](https://huggingface.co/OpenGVLab/InternVL3 - 14B)
InternVL3 - 38B	[InternViT - 6B - 448px - V2_5](https://huggingface.co/OpenGVLab/InternViT - 6B - 448px - V2_5)	[Qwen2.5 - 32B](https://huggingface.co/Qwen/Qwen2.5 - 32B)	[🤗 link](https://huggingface.co/OpenGVLab/InternVL3 - 38B)
InternVL3 - 78B	[InternViT - 6B - 448px - V2_5](https://huggingface.co/OpenGVLab/InternViT - 6B - 448px - V2_5)	[Qwen2.5 - 72B](https://huggingface.co/Qwen/Qwen2.5 - 72B)	[🤗 link](https://huggingface.co/OpenGVLab/InternVL3 - 78B)

模型架構

[InternVL3](https://internvl.github.io/blog/2025 - 04 - 11 - InternVL - 3/) 保留了與 [InternVL 2.5](https://internvl.github.io/blog/2024 - 12 - 05 - InternVL - 2.5/) 及其前身 InternVL 1.5 和 2.0 相同的模型架構，遵循 “ViT - MLP - LLM” 範式。在新版本中，使用隨機初始化的 MLP 投影器，將新的增量預訓練 InternViT 與多種預訓練 LLM（包括 InternLM 3 和 Qwen 2.5）集成。

與之前版本一樣，應用了像素重排操作，將視覺標記數量減少到原來的四分之一。此外，採用了與 InternVL 1.5 類似的動態分辨率策略，將圖像分割成 448×448 像素的圖塊。從 InternVL 2.0 開始的關鍵區別在於，還增加了對多圖像和視頻數據的支持。

值得注意的是，在 InternVL3 中集成了 可變視覺位置編碼 (V2PE)，它為視覺標記使用更小、更靈活的位置增量。得益於 V2PE，InternVL3 相比其前身表現出更好的長上下文理解能力。

訓練策略

原生多模態預訓練

提出了 原生多模態預訓練 方法，將語言和視覺學習整合到一個預訓練階段。與先訓練純語言模型，然後使其適應處理其他模態的標準範式不同，該方法將多模態數據（如圖文、視頻文本或圖文交錯序列）與大規模文本語料交織在一起。這種統一的訓練方案允許模型同時學習語言和多模態表示，最終增強其處理視覺 - 語言任務的能力，而無需單獨的對齊或橋接模塊。更多細節請參閱 我們的論文。

監督微調

在這個階段，InternVL2.5 中提出的隨機 JPEG 壓縮、平方損失重新加權和多模態數據打包技術也應用於 InternVL3 系列。與 InternVL2.5 相比，InternVL3 的 SFT 階段的主要進步在於使用了更高質量和更多樣化的訓練數據。具體來說，進一步擴展了工具使用、3D 場景理解、GUI 操作、長上下文任務、視頻理解、科學圖表、創意寫作和多模態推理的訓練樣本。

混合偏好優化

在預訓練和 SFT 期間，模型在先前真實標記的條件下預測下一個標記。然而，在推理期間，模型根據自己的先前輸出預測每個標記。真實標記和模型預測標記之間的這種差異引入了分佈偏移，這可能會損害模型的思維鏈 (CoT) 推理能力。為了緩解這個問題，採用了 MPO，它引入了來自正樣本和負樣本的額外監督，以使模型響應分佈與真實分佈對齊，從而提高推理性能。具體來說，MPO 的訓練目標是偏好損失 \(\mathcal{L}{\text{p}}\)、質量損失 \(\mathcal{L}{\text{q}}\) 和生成損失 \(\mathcal{L}_{\text{g}}\) 的組合，可以表述為：

$$ \mathcal{L}=w_{p}\cdot\mathcal{L}{\text{p}} + w{q}\cdot\mathcal{L}{\text{q}} + w{g}\cdot\mathcal{L}_{\text{g}}, $$

其中 \(w_{*}\) 表示分配給每個損失組件的權重。有關 MPO 的更多細節，請參閱 我們的論文。

測試時縮放

測試時縮放已被證明是增強 LLM 和 MLLM 推理能力的有效方法。在這項工作中，使用了 Best - of - N 評估策略，並採用 [VisualPRM - 8B](https://huggingface.co/OpenGVLab/VisualPRM - 8B) 作為評估模型，為推理和數學評估選擇最佳響應。

多模態能力評估

多模態推理和數學


OCR、圖表和文檔理解


多圖像和現實世界理解


綜合多模態和幻覺評估


視覺定位


多模態多語言理解


視頻理解


GUI 定位


空間推理


語言能力評估

將 InternVL3 與 Qwen2.5 Chat 模型進行比較，其對應的預訓練基礎模型用於初始化 InternVL3 中的語言組件。得益於原生多模態預訓練，InternVL3 系列在整體文本性能上比 Qwen2.5 系列更好。

請注意，Qwen2.5 系列的評估分數可能與官方報告的不同，因為在所有數據集上採用了表中提供的提示版本進行 OpenCompass 評估。


消融實驗

原生多模態預訓練

在保持 InternVL2 - 8B 模型的架構、初始化參數和訓練數據完全不變的情況下進行實驗。傳統上，InternVL2 - 8B 採用的訓練管道是先進行 MLP 預熱階段進行特徵對齊，然後進行指令微調階段。在實驗中，用原生多模態預訓練過程取代了傳統的 MLP 預熱階段。這種修改隔離了原生多模態預訓練對模型整體多模態能力的貢獻。

下圖中的評估結果表明，採用原生多模態預訓練的模型在大多數基準測試中的性能與經過完整多階段訓練的 InternVL2 - 8B 基線相當。此外，在更高質量數據上進行指令微調後，模型在評估的多模態任務中表現出進一步的性能提升。這些發現強調了原生多模態預訓練在賦予 MLLM 強大多模態能力方面的效率。


混合偏好優化

如下表所示，與未使用 MPO 進行微調的模型相比，使用 MPO 進行微調的模型在七個多模態推理基準測試中表現出更出色的推理性能。具體來說，InternVL3 - 78B 和 InternVL3 - 38B 分別比其對應模型高出 4.1 和 4.5 分。值得注意的是，用於 MPO 的訓練數據是用於 SFT 的訓練數據的子集，這表明性能提升主要源於訓練算法而非訓練數據。


可變視覺位置編碼

如下表所示，引入 V2PE 導致大多數評估指標的性能顯著提升。此外，通過改變位置增量 \( \delta \) 進行的消融實驗表明，即使對於主要涉及傳統上下文的任務，相對較小的 \( \delta \) 值也能實現最佳性能。這些發現為未來改進 MLLM 中視覺標記的位置編碼策略提供了重要見解。


🔧 技術細節

訓練策略

• 原生多模態預訓練：將語言和視覺學習整合到一個預訓練階段，通過交織多模態數據和大規模文本語料，使模型同時學習語言和多模態表示。
• 監督微調：使用隨機 JPEG 壓縮、平方損失重新加權和多模態數據打包技術，採用更高質量和更多樣化的訓練數據。
• 混合偏好優化：引入正樣本和負樣本的額外監督，使模型響應分佈與真實分佈對齊，提高推理性能。
• 測試時縮放：採用 Best - of - N 評估策略，使用 [VisualPRM - 8B](https://huggingface.co/OpenGVLab/VisualPRM - 8B) 作為評估模型，選擇最佳響應。

模型架構改進

• 集成 V2PE：在 InternVL3 中集成 可變視覺位置編碼 (V2PE)，使用更小、更靈活的位置增量，提升長上下文理解能力。
• 支持多圖像和視頻數據：從 InternVL 2.0 開始增加對多圖像和視頻數據的支持。

📦 安裝指南

LMDeploy

LMDeploy 是一個用於壓縮、部署和服務 LLM 和 VLM 的工具包。

# 如果 lmdeploy<0.7.3，需要顯式設置 chat_template_config=ChatTemplateConfig(model_name='internvl2_5')
pip install lmdeploy>=0.7.3

💻 使用示例

LMDeploy 示例

“Hello, world” 示例

from lmdeploy import pipeline, TurbomindEngineConfig, ChatTemplateConfig
from lmdeploy.vl import load_image

model = 'OpenGVLab/InternVL3-9B'
image = load_image('https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/tests/data/tiger.jpeg')
pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=16384, tp=1), chat_template_config=ChatTemplateConfig(model_name='internvl2_5'))
response = pipe(('describe this image', image))
print(response.text)

多圖像推理

from lmdeploy import pipeline, TurbomindEngineConfig, ChatTemplateConfig
from lmdeploy.vl import load_image
from lmdeploy.vl.constants import IMAGE_TOKEN

model = 'OpenGVLab/InternVL3-9B'
pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=16384, tp=1), chat_template_config=ChatTemplateConfig(model_name='internvl2_5'))

image_urls=[
    'https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/human-pose.jpg',
    'https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/det.jpg'
]

images = [load_image(img_url) for img_url in image_urls]
# 對圖像進行編號有助於多圖像對話
response = pipe((f'Image-1: {IMAGE_TOKEN}\nImage-2: {IMAGE_TOKEN}\ndescribe these two images', images))
print(response.text)

批量提示推理

from lmdeploy import pipeline, TurbomindEngineConfig, ChatTemplateConfig
from lmdeploy.vl import load_image

model = 'OpenGVLab/InternVL3-9B'
pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=16384, tp=1), chat_template_config=ChatTemplateConfig(model_name='internvl2_5'))

image_urls=[
    "https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/human-pose.jpg",
    "https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/det.jpg"
]
prompts = [('describe this image', load_image(img_url)) for img_url in image_urls]
response = pipe(prompts)
print(response)

多輪對話

from lmdeploy import pipeline, TurbomindEngineConfig, GenerationConfig, ChatTemplateConfig
from lmdeploy.vl import load_image

model = 'OpenGVLab/InternVL3-9B'
pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=16384, tp=1), chat_template_config=ChatTemplateConfig(model_name='internvl2_5'))

image = load_image('https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/human-pose.jpg')
gen_config = GenerationConfig(top_k=40, top_p=0.8, temperature=0.8)
sess = pipe.chat(('describe this image', image), gen_config=gen_config)
print(sess.response.text)
sess = pipe.chat('What is the woman doing?', session=sess, gen_config=gen_config)
print(sess.response.text)

服務部署

lmdeploy serve api_server OpenGVLab/InternVL3-9B --chat-template internvl2_5 --server-port 23333 --tp 1

使用 OpenAI 風格的接口，需要安裝 OpenAI：

pip install openai

然後使用以下代碼進行 API 調用：

from openai import OpenAI

client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')
model_name = client.models.list().data[0].id
response = client.chat.completions.create(
    model=model_name,
    messages=[{
        'role':
        'user',
        'content': [{
            'type': 'text',
            'text': 'describe this image',
        }, {
            'type': 'image_url',
            'image_url': {
                'url':
                'https://modelscope.oss-cn-beijing.aliyuncs.com/resource/tiger.jpeg',
            },
        }],
    }],
    temperature=0.8,
    top_p=0.8)
print(response)

📄 許可證

本項目遵循 MIT 許可證發佈。本項目使用預訓練的 Qwen2.5 作為組件，該組件遵循 Qwen 許可證。

引用

如果您在研究中發現本項目有用，請考慮引用：

@article{chen2024expanding,
  title={Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling},
  author={Chen, Zhe and Wang, Weiyun and Cao, Yue and Liu, Yangzhou and Gao, Zhangwei and Cui, Erfei and Zhu, Jinguo and Ye, Shenglong and Tian, Hao and Liu, Zhaoyang and others},
  journal={arXiv preprint arXiv:2412.05271},
  year={2024}
}
@article{wang2024mpo,
  title={Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization},
  author={Wang, Weiyun and Chen, Zhe and Wang, Wenhai and Cao, Yue and Liu, Yangzhou and Gao, Zhangwei and Zhu, Jinguo and Zhu, Xizhou and Lu, Lewei and Qiao, Yu and Dai, Jifeng},
  journal={arXiv preprint arXiv:2411.10442},
  year={2024}
}
@article{chen2024far,
  title={How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites},
  author={Chen, Zhe and Wang, Weiyun and Tian, Hao and Ye, Shenglong and Gao, Zhangwei and Cui, Erfei and Tong, Wenwen and Hu, Kongzhi and Luo, Jiapeng and Ma, Zheng and others},
  journal={arXiv preprint arXiv:2404.16821},
  year={2024}
}
@inproceedings{chen2024internvl,
  title={Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks},
  author={Chen, Zhe and Wu, Jiannan and Wang, Wenhai and Su, Weijie and Chen, Guo and Xing, Sen and Zhong, Muyan and Zhang, Qinglong and Zhu, Xizhou and Lu, Lewei and others},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={24185--24198},
  year={2024}
}