[Feat] Add llava qwen, llava mistral (#419)

Co-authored-by: Bo Li <drluodian@gmail.com>

examples/usage/llava/http_llama3_llava_test.py

@@ -0,0 +1,117 @@
+"""
+Usage:
+# Installing latest llava-next: pip install git+https://github.com/LLaVA-VL/LLaVA-NeXT.git
+# Installing latest sglang.
+
+# Endpoint Service CLI: 
+# python -m sglang.launch_server --model-path lmms-lab/llama3-llava-next-8b --tokenizer-path lmms-lab/llama3-llava-next-8b-tokenizer --port=30000 --host="127.0.0.1" --tp-size=4
+
+python3 http_llama3_llava_test.py
+
+Output:
+"Friends posing for a fun photo with a life-sized teddy bear, creating a playful and memorable moment."
+"""
+
+import argparse
+import asyncio
+import json
+import time
+import copy
+
+import aiohttp
+import requests
+
+from llava.conversation import (
+    default_conversation,
+    conv_templates,
+    SeparatorStyle,
+    conv_llava_llama_3,
+    conv_qwen,
+)
+
+
+async def send_request(url, data, delay=0):
+    await asyncio.sleep(delay)
+    async with aiohttp.ClientSession() as session:
+        async with session.post(url, json=data) as resp:
+            output = await resp.json()
+    return output
+
+
+async def test_concurrent(args):
+    url = f"{args.host}:{args.port}"
+
+    prompt = "<image>\nPlease generate caption towards this image."
+    conv_template = copy.deepcopy(conv_llava_llama_3)
+    conv_template.append_message(role="user", message=prompt)
+    prompt_with_template = conv_template.get_prompt()
+    response = []
+    for i in range(1):
+        response.append(
+            send_request(
+                url + "/generate",
+                {
+                    "text": prompt_with_template,
+                    "image_data": "https://farm4.staticflickr.com/3175/2653711032_804ff86d81_z.jpg",
+                    "sampling_params": {
+                        "max_new_tokens": 1024,
+                        "temperature": 0,
+                        "top_p": 1.0,
+                        "presence_penalty": 2,
+                        "frequency_penalty": 2,
+                        "stop": "<|eot_id|>",
+                    },
+                },
+            )
+        )
+
+    rets = await asyncio.gather(*response)
+    for ret in rets:
+        print(ret["text"])
+
+
+def test_streaming(args):
+    url = f"{args.host}:{args.port}"
+    prompt = "<image>\nPlease generate caption towards this image."
+    conv_template = copy.deepcopy(conv_llava_llama_3)
+    conv_template.append_message(role="user", message=prompt)
+    prompt_with_template = conv_template.get_prompt()
+    pload = {
+        "text": prompt_with_template,
+        "sampling_params": {
+            "max_new_tokens": 1024,
+            "temperature": 0,
+            "top_p": 1.0,
+            "presence_penalty": 2,
+            "frequency_penalty": 2,
+            "stop": "<|eot_id|>",
+        },
+        "image_data": "https://farm4.staticflickr.com/3175/2653711032_804ff86d81_z.jpg",
+        "stream": True,
+    }
+    response = requests.post(
+        url + "/generate",
+        json=pload,
+        stream=True,
+    )
+
+    prev = 0
+    for chunk in response.iter_lines(decode_unicode=False):
+        chunk = chunk.decode("utf-8")
+        if chunk and chunk.startswith("data:"):
+            if chunk == "data: [DONE]":
+                break
+            data = json.loads(chunk[5:].strip("\n"))
+            output = data["text"].strip()
+            print(output[prev:], end="", flush=True)
+            prev = len(output)
+    print("")
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--host", type=str, default="http://127.0.0.1")
+    parser.add_argument("--port", type=int, default=30000)
+    args = parser.parse_args()
+    asyncio.run(test_concurrent(args))
+    test_streaming(args)

examples/usage/llava/http_qwen_llava_test.py

@@ -0,0 +1,117 @@
+"""
+Usage:
+# Installing latest llava-next: pip install git+https://github.com/LLaVA-VL/LLaVA-NeXT.git
+# Installing latest sglang.
+
+# Endpoint Service CLI: 
+# python -m sglang.launch_server --model-path lmms-lab/llava-next-72b --tokenizer-path lmms-lab/llavanext-qwen-tokenizer --port=30000 --host="127.0.0.1" --tp-size=4
+
+python3 http_qwen_llava_test.py
+
+Output:
+"Two children pose with a large teddy bear, one holding a smaller stuffed bear, in a room with an American flag and potted plants."
+"""
+
+import argparse
+import asyncio
+import json
+import time
+import copy
+
+import aiohttp
+import requests
+
+from llava.conversation import (
+    default_conversation,
+    conv_templates,
+    SeparatorStyle,
+    conv_llava_llama_3,
+    conv_qwen,
+)
+
+
+async def send_request(url, data, delay=0):
+    await asyncio.sleep(delay)
+    async with aiohttp.ClientSession() as session:
+        async with session.post(url, json=data) as resp:
+            output = await resp.json()
+    return output
+
+
+async def test_concurrent(args):
+    url = f"{args.host}:{args.port}"
+
+    prompt = "<image>\nPlease generate caption towards this image."
+    conv_template = copy.deepcopy(conv_qwen)
+    conv_template.append_message(role="user", message=prompt)
+    prompt_with_template = conv_template.get_prompt()
+    response = []
+    for i in range(1):
+        response.append(
+            send_request(
+                url + "/generate",
+                {
+                    "text": prompt_with_template,
+                    "image_data": "https://farm4.staticflickr.com/3175/2653711032_804ff86d81_z.jpg",
+                    "sampling_params": {
+                        "max_new_tokens": 1024,
+                        "temperature": 0,
+                        "top_p": 1.0,
+                        "presence_penalty": 2,
+                        "frequency_penalty": 2,
+                        "stop": "<|im_end|>",
+                    },
+                },
+            )
+        )
+
+    rets = await asyncio.gather(*response)
+    for ret in rets:
+        print(ret["text"])
+
+
+def test_streaming(args):
+    url = f"{args.host}:{args.port}"
+    prompt = "<image>\nPlease generate caption towards this image."
+    conv_template = copy.deepcopy(conv_qwen)
+    conv_template.append_message(role="user", message=prompt)
+    prompt_with_template = conv_template.get_prompt()
+    pload = {
+        "text": prompt_with_template,
+        "sampling_params": {
+            "max_new_tokens": 1024,
+            "temperature": 0,
+            "top_p": 1.0,
+            "presence_penalty": 2,
+            "frequency_penalty": 2,
+            "stop": "<|im_end|>",
+        },
+        "image_data": "https://farm4.staticflickr.com/3175/2653711032_804ff86d81_z.jpg",
+        "stream": True,
+    }
+    response = requests.post(
+        url + "/generate",
+        json=pload,
+        stream=True,
+    )
+
+    prev = 0
+    for chunk in response.iter_lines(decode_unicode=False):
+        chunk = chunk.decode("utf-8")
+        if chunk and chunk.startswith("data:"):
+            if chunk == "data: [DONE]":
+                break
+            data = json.loads(chunk[5:].strip("\n"))
+            output = data["text"].strip()
+            print(output[prev:], end="", flush=True)
+            prev = len(output)
+    print("")
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--host", type=str, default="http://127.0.0.1")
+    parser.add_argument("--port", type=int, default=30000)
+    args = parser.parse_args()
+    # asyncio.run(test_concurrent(args))
+    test_streaming(args)

examples/usage/llava/srt_llava_next_test.py

@@ -0,0 +1,88 @@
+"""
+Usage: python3 srt_example_llava.py
+"""
+
+import sglang as sgl
+from sglang.srt.utils import load_image
+from sglang.lang.chat_template import get_chat_template
+
+from PIL import ImageFile
+ImageFile.LOAD_TRUNCATED_IMAGES = True  # Allow loading of truncated images
+
+@sgl.function
+def image_qa(s, image, question):
+    s += sgl.user(sgl.image(image) + question)
+    s += sgl.assistant(sgl.gen("answer"))
+
+
+def single():
+    image_url = "https://farm4.staticflickr.com/3175/2653711032_804ff86d81_z.jpg"
+    pil_image = load_image(image_url)
+    state = image_qa.run(image=pil_image, question="What is this?", max_new_tokens=512)
+    print(state["answer"], "\n")
+
+
+def stream():
+    image_url = "https://farm4.staticflickr.com/3175/2653711032_804ff86d81_z.jpg"
+    pil_image = load_image(image_url)
+    state = image_qa.run(
+        image=pil_image,
+        question="Please generate short caption for this image.",
+        max_new_tokens=512,
+        temperature=0,
+        stream=True,
+    )
+
+    for out in state.text_iter("answer"):
+        print(out, end="", flush=True)
+    print()
+
+
+def batch():
+    image_url = "https://farm4.staticflickr.com/3175/2653711032_804ff86d81_z.jpg"
+    pil_image = load_image(image_url)
+    states = image_qa.run_batch(
+        [
+            {"image": pil_image, "question": "What is this?"},
+            {"image": pil_image, "question": "What is this?"},
+        ],
+        max_new_tokens=512,
+    )
+    for s in states:
+        print(s["answer"], "\n")
+
+
+if __name__ == "__main__":
+    import multiprocessing as mp
+
+    mp.set_start_method("spawn", force=True)
+    runtime = sgl.Runtime(
+        model_path="lmms-lab/llama3-llava-next-8b",
+        tokenizer_path="lmms-lab/llama3-llava-next-8b-tokenizer",
+    )
+    runtime.endpoint.chat_template = get_chat_template("llama-3-instruct")
+    # runtime = sgl.Runtime(
+    #     model_path="lmms-lab/llava-next-72b",
+    #     tokenizer_path="lmms-lab/llavanext-qwen-tokenizer",
+    # )
+    # runtime.endpoint.chat_template = get_chat_template("chatml-llava")
+    sgl.set_default_backend(runtime)
+    print(f"chat template: {runtime.endpoint.chat_template.name}")
+
+    # Or you can use API models
+    # sgl.set_default_backend(sgl.OpenAI("gpt-4-vision-preview"))
+    # sgl.set_default_backend(sgl.VertexAI("gemini-pro-vision"))
+
+    # Run a single request
+    print("\n========== single ==========\n")
+    single()
+
+    # Stream output
+    print("\n========== stream ==========\n")
+    stream()
+
+    # Run a batch of requests
+    print("\n========== batch ==========\n")
+    batch()
+
+    runtime.shutdown()

python/sglang/srt/models/llava.py

@@ -328,4 +328,4 @@ def monkey_path_clip_vision_embed_forward():
     )
 
 
-EntryClass = LlavaLlamaForCausalLM
+EntryClass = LlavaLlamaForCausalLM

python/sglang/srt/models/llava_mistral.py

@@ -0,0 +1,347 @@
+"""Inference-only LLaVa model compatible with HuggingFace weights."""
+
+from typing import List, Optional
+
+import numpy as np
+import torch
+from torch import nn
+from transformers import CLIPVisionModel, LlavaConfig, CLIPVisionConfig, MistralConfig
+from transformers.models.llava.modeling_llava import LlavaMultiModalProjector
+from vllm.model_executor.layers.quantization.base_config import QuantizationConfig
+from sglang.srt.weight_utils import (
+    default_weight_loader,
+    hf_model_weights_iterator,
+)
+
+from sglang.srt.managers.router.infer_batch import ForwardMode
+from sglang.srt.managers.router.model_runner import InputMetadata
+from sglang.srt.mm_utils import (
+    get_anyres_image_grid_shape,
+    unpad_image,
+    unpad_image_shape,
+)
+from sglang.srt.models.mistral import MistralForCausalLM
+
+
+class LlavaMistralForCausalLM(nn.Module):
+    def __init__(
+        self,
+        config: LlavaConfig,
+        quant_config: Optional[QuantizationConfig] = None,
+    ) -> None:
+        super().__init__()
+        self.config = config
+        self.vision_tower = None
+        if getattr(self.config, "vision_config", None) is None:
+            self.config.vision_config = CLIPVisionConfig(self.config.mm_vision_tower)
+
+        if getattr(self.config, "text_config", None) is None:
+            self.config.text_config = MistralConfig(self.config._name_or_path)
+
+        self.config.vision_config.hidden_size = config.mm_hidden_size
+        self.config.text_config.hidden_size = config.hidden_size
+
+        if getattr(self.config, "projector_hidden_act", None) is None:
+            self.config.projector_hidden_act = "gelu"
+
+        if getattr(self.config, "image_token_index", None) is None:
+            self.config.image_token_index = 32000
+
+        self.multi_modal_projector = LlavaMultiModalProjector(config)
+        self.language_model = MistralForCausalLM(config, quant_config=quant_config)
+        if "unpad" in getattr(config, "mm_patch_merge_type", ""):
+            self.language_model.model.image_newline = nn.Parameter(
+                torch.empty(config.text_config.hidden_size, dtype=torch.float16)
+            )
+
+    def pad_input_ids(self, input_ids, pad_value, pt_shape=None, image_size=None):
+        new_image_feature_len = self.image_feature_len
+        # now only support spatial_unpad + anyres
+        if self.mm_patch_merge_type.startswith("spatial"):
+            height = width = self.num_patches_per_side
+            if pt_shape[0] > 1:
+                if self.image_aspect_ratio == "anyres":
+                    num_patch_width, num_patch_height = get_anyres_image_grid_shape(
+                        image_size,
+                        self.image_grid_pinpoints,
+                        self.vision_tower.config.image_size,
+                    )
+                if "unpad" in self.mm_patch_merge_type:
+                    h = num_patch_height * height
+                    w = num_patch_width * width
+                    new_h, new_w = unpad_image_shape(h, w, image_size)
+                    new_image_feature_len += new_h * (new_w + 1)
+
+        pad_ids = pad_value * (
+            (new_image_feature_len + len(pad_value)) // len(pad_value)
+        )
+        offset = input_ids.index(self.config.image_token_index)
+        # old_len + pad_len - 1, because we need to remove image_token_id
+        new_input_ids = (
+            input_ids[:offset]
+            + pad_ids[:new_image_feature_len]
+            + input_ids[offset + 1 :]
+        )
+        return new_input_ids, offset
+
+    def encode_images(self, pixel_values: torch.Tensor) -> torch.Tensor:
+        image_outputs = self.vision_tower(pixel_values, output_hidden_states=True)
+        # NOTE: This is not memory efficient. (output_hidden_states=True) will save all the hidden stated.
+
+        selected_image_feature = image_outputs.hidden_states[self.vision_feature_layer]
+        if self.vision_feature_select_strategy in ["default", "patch"]:
+            selected_image_feature = selected_image_feature[:, 1:]
+        elif self.vision_feature_select_strategy == "full":
+            selected_image_feature = selected_image_feature
+        else:
+            raise ValueError(
+                f"Unexpected select feature strategy: {self.config.vision_feature_select_strategy}"
+            )
+        image_features = self.multi_modal_projector(selected_image_feature)
+
+        return image_features
+
+    def forward(
+        self,
+        input_ids: torch.LongTensor,
+        positions: torch.Tensor,
+        input_metadata: InputMetadata,
+        pixel_values: Optional[List[Optional[np.array]]] = None,
+        image_sizes: Optional[List[List[int]]] = None,
+        image_offsets: Optional[List[int]] = None,
+    ) -> torch.Tensor:
+        if input_metadata.forward_mode == ForwardMode.EXTEND:
+            bs = input_metadata.batch_size
+
+            # Embed text input
+            input_embeds = self.language_model.model.embed_tokens(input_ids)
+
+            # Embed vision input
+            need_vision = (
+                (positions[input_metadata.extend_start_loc] < self.image_feature_len)
+                .cpu()
+                .numpy()
+            )
+            # FIXME: We need to substract the length of the system prompt
+            has_pixel = np.array([pixel_values[i] is not None for i in range(bs)])
+            need_vision = need_vision & has_pixel
+
+            if need_vision.any():
+                pixel_values = [pixel_values[i] for i in range(bs) if need_vision[i]]
+                image_sizes = [image_sizes[i] for i in range(bs) if need_vision[i]]
+
+                ########## Encode Image ########
+
+                if pixel_values[0].ndim == 4:
+                    # llava-hd: BS, num_patch, C=3, H=336, W=336, num_patch obtained from process_images
+                    np.concatenate(pixel_values, axis=0)
+                    # ndim=4
+                    concat_images = torch.tensor(
+                        np.concatenate(pixel_values, axis=0),
+                        device=self.vision_tower.device,
+                    )
+                    image_features = self.encode_images(concat_images)
+                    split_sizes = [image.shape[0] for image in pixel_values]
+                    image_features = torch.split(image_features, split_sizes, dim=0)
+                    # hd image_features: BS, num_patch, 576, 4096
+                else:
+                    # normal pixel: BS, C=3, H=336, W=336
+                    pixel_values = torch.tensor(
+                        np.array(pixel_values), device=self.vision_tower.device
+                    )
+                    image_features = self.encode_images(pixel_values)
+                    # image_features: BS, 576, 4096
+
+                if self.mm_patch_merge_type.startswith("spatial"):
+                    new_image_features = []
+                    for image_idx, image_feature in enumerate(image_features):
+                        if image_feature.shape[0] > 1:
+                            base_image_feature = image_feature[0]
+                            image_feature = image_feature[1:]
+                            height = width = self.num_patches_per_side
+                            assert height * width == base_image_feature.shape[0]
+                            if self.image_aspect_ratio == "anyres":
+                                (
+                                    num_patch_width,
+                                    num_patch_height,
+                                ) = get_anyres_image_grid_shape(
+                                    image_sizes[image_idx],
+                                    self.image_grid_pinpoints,
+                                    self.vision_tower.config.image_size,
+                                )
+                                image_feature = image_feature.view(
+                                    num_patch_height, num_patch_width, height, width, -1
+                                )
+                            else:
+                                raise NotImplementedError()
+                            if "unpad" in self.mm_patch_merge_type:
+                                image_feature = image_feature.permute(
+                                    4, 0, 2, 1, 3
+                                ).contiguous()
+                                image_feature = image_feature.flatten(1, 2).flatten(
+                                    2, 3
+                                )
+                                image_feature = unpad_image(
+                                    image_feature, image_sizes[image_idx]
+                                )
+                                image_feature = torch.cat(
+                                    (
+                                        image_feature,
+                                        self.language_model.model.image_newline[
+                                            :, None, None
+                                        ].expand(*image_feature.shape[:-1], 1),
+                                    ),
+                                    dim=-1,
+                                )
+                                image_feature = image_feature.flatten(1, 2).transpose(
+                                    0, 1
+                                )
+                            else:
+                                image_feature = image_feature.permute(
+                                    0, 2, 1, 3, 4
+                                ).contiguous()
+                                image_feature = image_feature.flatten(0, 3)
+                            image_feature = torch.cat(
+                                (base_image_feature, image_feature), dim=0
+                            )
+                        else:
+                            image_feature = image_feature[0]
+                            if "unpad" in self.mm_patch_merge_type:
+                                image_feature = torch.cat(
+                                    (
+                                        image_feature,
+                                        self.language_model.model.image_newline[None],
+                                    ),
+                                    dim=0,
+                                )
+                        new_image_features.append(image_feature)
+                    image_features = new_image_features
+
+                extend_start_loc_cpu = input_metadata.extend_start_loc.cpu().numpy()
+                pt = 0
+                for i in range(bs):
+                    if not need_vision[i]:
+                        continue
+
+                    start_idx = extend_start_loc_cpu[i]
+                    pad_len, pad_dim = image_features[pt].shape  # 576, 4096
+                    dim = input_embeds.shape[1]
+                    assert (
+                        pad_dim == dim
+                    ), "invalid pad_dim={}, input_embed_dim={}!".format(pad_dim, dim)
+                    # Fill in the placeholder for the image
+                    try:
+                        input_embeds[
+                            start_idx
+                            + image_offsets[i] : start_idx
+                            + image_offsets[i]
+                            + pad_len
+                        ] = image_features[pt]
+                    except RuntimeError as e:
+                        print(f"RuntimeError in llava image encoding: {e}")
+                        print(input_embeds.shape)
+                        print(start_idx, image_offsets[i])
+                    pt += 1
+
+            return self.language_model(
+                input_ids, positions, input_metadata, input_embeds=input_embeds
+            )
+        elif input_metadata.forward_mode == ForwardMode.DECODE:
+            return self.language_model(input_ids, positions, input_metadata)
+
+    def load_weights(
+        self,
+        model_name_or_path: str,
+        cache_dir: Optional[str] = None,
+        load_format: str = "auto",
+        revision: Optional[str] = None,
+    ):
+        # load clip vision model by cfg['mm_vision_tower']:
+        #   huggingface_name or path_of_clip_relative_to_llava_model_dir
+        vision_path = self.config.mm_vision_tower
+        self.vision_tower = CLIPVisionModel.from_pretrained(
+            vision_path, torch_dtype=torch.float16
+        ).cuda()
+        self.vision_tower.eval()
+
+        self.vision_feature_layer = self.config.mm_vision_select_layer
+        self.vision_feature_select_strategy = self.config.mm_vision_select_feature
+        self.image_size = self.vision_tower.config.image_size
+        self.patch_size = self.vision_tower.config.patch_size
+
+        self.mm_patch_merge_type = getattr(self.config, "mm_patch_merge_type", "flat")
+        self.image_aspect_ratio = getattr(self.config, "image_aspect_ratio", "square")
+        self.image_grid_pinpoints = getattr(self.config, "image_grid_pinpoints", None)
+
+        self.image_feature_len = int((self.image_size / self.patch_size) ** 2)
+        if self.vision_feature_select_strategy == "patch":
+            pass
+        elif self.vision_feature_select_strategy == "cls_patch":
+            self.image_feature_len += 1
+        else:
+            raise ValueError(f"Unexpected select feature: {self.select_feature}")
+
+        # load mm_projector
+        projector_weights = {
+            "model.mm_projector.0": "multi_modal_projector.linear_1",
+            "model.mm_projector.2": "multi_modal_projector.linear_2",
+            "model.vision_tower.vision_tower": "vision_tower",  # Update the vision tower weights if we find them in the checkpoint (it may be finetuned).
+        }
+        params_dict = dict(self.named_parameters())
+        for name, loaded_weight in hf_model_weights_iterator(
+            model_name_or_path, cache_dir, load_format, revision
+        ):
+            # FIXME: why projector weights read two times?
+            if "projector" in name or "vision_tower" in name:
+                for weight_name, param_name in projector_weights.items():
+                    if weight_name in name:
+                        name = name.replace(weight_name, param_name)
+                param = params_dict[name]
+                weight_loader = getattr(param, "weight_loader", default_weight_loader)
+                weight_loader(param, loaded_weight)
+
+        # load language model
+        self.language_model.load_weights(
+            model_name_or_path, cache_dir, load_format, revision
+        )
+
+        monkey_path_clip_vision_embed_forward()
+
+    @property
+    def num_patches_per_side(self):
+        return self.image_size // self.patch_size
+
+
+first_call = True
+
+
+def clip_vision_embed_forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
+    batch_size = pixel_values.shape[0]
+
+    # Move this conv layer to CPU to avoid a bug in torch >= 2.1 on A10G.
+    global first_call
+    if first_call:
+        self.patch_embedding.cpu().float()
+        first_call = False
+    pixel_values = pixel_values.to(dtype=torch.float32, device="cpu")
+    patch_embeds = self.patch_embedding(pixel_values).cuda().half()
+
+    patch_embeds = patch_embeds.flatten(2).transpose(1, 2)
+
+    class_embeds = self.class_embedding.expand(batch_size, 1, -1)
+    embeddings = torch.cat([class_embeds, patch_embeds], dim=1)
+    embeddings = embeddings + self.position_embedding(self.position_ids)
+    return embeddings
+
+
+def monkey_path_clip_vision_embed_forward():
+    import transformers
+
+    setattr(
+        transformers.models.clip.modeling_clip.CLIPVisionEmbeddings,
+        "forward",
+        clip_vision_embed_forward,
+    )
+
+
+EntryClass = LlavaMistralForCausalLM

python/sglang/srt/models/llava_qwen.py

@@ -0,0 +1,347 @@
+"""Inference-only LLaVa model compatible with HuggingFace weights."""
+
+from typing import List, Optional
+
+import numpy as np
+import torch
+from torch import nn
+from transformers import CLIPVisionModel, LlavaConfig, CLIPVisionConfig, Qwen2Config
+from transformers.models.llava.modeling_llava import LlavaMultiModalProjector
+from vllm.model_executor.layers.quantization.base_config import QuantizationConfig
+from sglang.srt.weight_utils import (
+    default_weight_loader,
+    hf_model_weights_iterator,
+)
+
+from sglang.srt.managers.router.infer_batch import ForwardMode
+from sglang.srt.managers.router.model_runner import InputMetadata
+from sglang.srt.mm_utils import (
+    get_anyres_image_grid_shape,
+    unpad_image,
+    unpad_image_shape,
+)
+from sglang.srt.models.qwen2 import Qwen2ForCausalLM
+
+
+class LlavaQwenForCausalLM(nn.Module):
+    def __init__(
+        self,
+        config: LlavaConfig,
+        quant_config: Optional[QuantizationConfig] = None,
+    ) -> None:
+        super().__init__()
+        self.config = config
+        self.vision_tower = None
+        if getattr(self.config, "vision_config", None) is None:
+            self.config.vision_config = CLIPVisionConfig(self.config.mm_vision_tower)
+
+        if getattr(self.config, "text_config", None) is None:
+            self.config.text_config = Qwen2Config(self.config._name_or_path)
+
+        self.config.vision_config.hidden_size = config.mm_hidden_size
+        self.config.text_config.hidden_size = config.hidden_size
+
+        if getattr(self.config, "projector_hidden_act", None) is None:
+            self.config.projector_hidden_act = "gelu"
+
+        if getattr(self.config, "image_token_index", None) is None:
+            self.config.image_token_index = 151646
+
+        self.multi_modal_projector = LlavaMultiModalProjector(config)
+        self.language_model = Qwen2ForCausalLM(config, quant_config=quant_config)
+        if "unpad" in getattr(config, "mm_patch_merge_type", ""):
+            self.language_model.model.image_newline = nn.Parameter(
+                torch.empty(config.text_config.hidden_size, dtype=torch.float16)
+            )
+
+    def pad_input_ids(self, input_ids, pad_value, pt_shape=None, image_size=None):
+        new_image_feature_len = self.image_feature_len
+        # now only support spatial_unpad + anyres
+        if self.mm_patch_merge_type.startswith("spatial"):
+            height = width = self.num_patches_per_side
+            if pt_shape[0] > 1:
+                if self.image_aspect_ratio == "anyres":
+                    num_patch_width, num_patch_height = get_anyres_image_grid_shape(
+                        image_size,
+                        self.image_grid_pinpoints,
+                        self.vision_tower.config.image_size,
+                    )
+                if "unpad" in self.mm_patch_merge_type:
+                    h = num_patch_height * height
+                    w = num_patch_width * width
+                    new_h, new_w = unpad_image_shape(h, w, image_size)
+                    new_image_feature_len += new_h * (new_w + 1)
+
+        pad_ids = pad_value * (
+            (new_image_feature_len + len(pad_value)) // len(pad_value)
+        )
+        offset = input_ids.index(self.config.image_token_index)
+        # old_len + pad_len - 1, because we need to remove image_token_id
+        new_input_ids = (
+            input_ids[:offset]
+            + pad_ids[:new_image_feature_len]
+            + input_ids[offset + 1 :]
+        )
+        return new_input_ids, offset
+
+    def encode_images(self, pixel_values: torch.Tensor) -> torch.Tensor:
+        image_outputs = self.vision_tower(pixel_values, output_hidden_states=True)
+        # NOTE: This is not memory efficient. (output_hidden_states=True) will save all the hidden stated.
+
+        selected_image_feature = image_outputs.hidden_states[self.vision_feature_layer]
+        if self.vision_feature_select_strategy in ["default", "patch"]:
+            selected_image_feature = selected_image_feature[:, 1:]
+        elif self.vision_feature_select_strategy == "full":
+            selected_image_feature = selected_image_feature
+        else:
+            raise ValueError(
+                f"Unexpected select feature strategy: {self.config.vision_feature_select_strategy}"
+            )
+        image_features = self.multi_modal_projector(selected_image_feature)
+
+        return image_features
+
+    def forward(
+        self,
+        input_ids: torch.LongTensor,
+        positions: torch.Tensor,
+        input_metadata: InputMetadata,
+        pixel_values: Optional[List[Optional[np.array]]] = None,
+        image_sizes: Optional[List[List[int]]] = None,
+        image_offsets: Optional[List[int]] = None,
+    ) -> torch.Tensor:
+        if input_metadata.forward_mode == ForwardMode.EXTEND:
+            bs = input_metadata.batch_size
+
+            # Embed text input
+            input_embeds = self.language_model.model.embed_tokens(input_ids)
+
+            # Embed vision input
+            need_vision = (
+                (positions[input_metadata.extend_start_loc] < self.image_feature_len)
+                .cpu()
+                .numpy()
+            )
+            # FIXME: We need to substract the length of the system prompt
+            has_pixel = np.array([pixel_values[i] is not None for i in range(bs)])
+            need_vision = need_vision & has_pixel
+
+            if need_vision.any():
+                pixel_values = [pixel_values[i] for i in range(bs) if need_vision[i]]
+                image_sizes = [image_sizes[i] for i in range(bs) if need_vision[i]]
+
+                ########## Encode Image ########
+
+                if pixel_values[0].ndim == 4:
+                    # llava-hd: BS, num_patch, C=3, H=336, W=336, num_patch obtained from process_images
+                    np.concatenate(pixel_values, axis=0)
+                    # ndim=4
+                    concat_images = torch.tensor(
+                        np.concatenate(pixel_values, axis=0),
+                        device=self.vision_tower.device,
+                    )
+                    image_features = self.encode_images(concat_images)
+                    split_sizes = [image.shape[0] for image in pixel_values]
+                    image_features = torch.split(image_features, split_sizes, dim=0)
+                    # hd image_features: BS, num_patch, 576, 4096
+                else:
+                    # normal pixel: BS, C=3, H=336, W=336
+                    pixel_values = torch.tensor(
+                        np.array(pixel_values), device=self.vision_tower.device
+                    )
+                    image_features = self.encode_images(pixel_values)
+                    # image_features: BS, 576, 4096
+
+                if self.mm_patch_merge_type.startswith("spatial"):
+                    new_image_features = []
+                    for image_idx, image_feature in enumerate(image_features):
+                        if image_feature.shape[0] > 1:
+                            base_image_feature = image_feature[0]
+                            image_feature = image_feature[1:]
+                            height = width = self.num_patches_per_side
+                            assert height * width == base_image_feature.shape[0]
+                            if self.image_aspect_ratio == "anyres":
+                                (
+                                    num_patch_width,
+                                    num_patch_height,
+                                ) = get_anyres_image_grid_shape(
+                                    image_sizes[image_idx],
+                                    self.image_grid_pinpoints,
+                                    self.vision_tower.config.image_size,
+                                )
+                                image_feature = image_feature.view(
+                                    num_patch_height, num_patch_width, height, width, -1
+                                )
+                            else:
+                                raise NotImplementedError()
+                            if "unpad" in self.mm_patch_merge_type:
+                                image_feature = image_feature.permute(
+                                    4, 0, 2, 1, 3
+                                ).contiguous()
+                                image_feature = image_feature.flatten(1, 2).flatten(
+                                    2, 3
+                                )
+                                image_feature = unpad_image(
+                                    image_feature, image_sizes[image_idx]
+                                )
+                                image_feature = torch.cat(
+                                    (
+                                        image_feature,
+                                        self.language_model.model.image_newline[
+                                            :, None, None
+                                        ].expand(*image_feature.shape[:-1], 1),
+                                    ),
+                                    dim=-1,
+                                )
+                                image_feature = image_feature.flatten(1, 2).transpose(
+                                    0, 1
+                                )
+                            else:
+                                image_feature = image_feature.permute(
+                                    0, 2, 1, 3, 4
+                                ).contiguous()
+                                image_feature = image_feature.flatten(0, 3)
+                            image_feature = torch.cat(
+                                (base_image_feature, image_feature), dim=0
+                            )
+                        else:
+                            image_feature = image_feature[0]
+                            if "unpad" in self.mm_patch_merge_type:
+                                image_feature = torch.cat(
+                                    (
+                                        image_feature,
+                                        self.language_model.model.image_newline[None],
+                                    ),
+                                    dim=0,
+                                )
+                        new_image_features.append(image_feature)
+                    image_features = new_image_features
+
+                extend_start_loc_cpu = input_metadata.extend_start_loc.cpu().numpy()
+                pt = 0
+                for i in range(bs):
+                    if not need_vision[i]:
+                        continue
+
+                    start_idx = extend_start_loc_cpu[i]
+                    pad_len, pad_dim = image_features[pt].shape  # 576, 4096
+                    dim = input_embeds.shape[1]
+                    assert (
+                        pad_dim == dim
+                    ), "invalid pad_dim={}, input_embed_dim={}!".format(pad_dim, dim)
+                    # Fill in the placeholder for the image
+                    try:
+                        input_embeds[
+                            start_idx
+                            + image_offsets[i] : start_idx
+                            + image_offsets[i]
+                            + pad_len
+                        ] = image_features[pt]
+                    except RuntimeError as e:
+                        print(f"RuntimeError in llava image encoding: {e}")
+                        print(input_embeds.shape)
+                        print(start_idx, image_offsets[i])
+                    pt += 1
+
+            return self.language_model(
+                input_ids, positions, input_metadata, input_embeds=input_embeds
+            )
+        elif input_metadata.forward_mode == ForwardMode.DECODE:
+            return self.language_model(input_ids, positions, input_metadata)
+
+    def load_weights(
+        self,
+        model_name_or_path: str,
+        cache_dir: Optional[str] = None,
+        load_format: str = "auto",
+        revision: Optional[str] = None,
+    ):
+        # load clip vision model by cfg['mm_vision_tower']:
+        #   huggingface_name or path_of_clip_relative_to_llava_model_dir
+        vision_path = self.config.mm_vision_tower
+        self.vision_tower = CLIPVisionModel.from_pretrained(
+            vision_path, torch_dtype=torch.float16
+        ).cuda()
+        self.vision_tower.eval()
+
+        self.vision_feature_layer = self.config.mm_vision_select_layer
+        self.vision_feature_select_strategy = self.config.mm_vision_select_feature
+        self.image_size = self.vision_tower.config.image_size
+        self.patch_size = self.vision_tower.config.patch_size
+
+        self.mm_patch_merge_type = getattr(self.config, "mm_patch_merge_type", "flat")
+        self.image_aspect_ratio = getattr(self.config, "image_aspect_ratio", "square")
+        self.image_grid_pinpoints = getattr(self.config, "image_grid_pinpoints", None)
+
+        self.image_feature_len = int((self.image_size / self.patch_size) ** 2)
+        if self.vision_feature_select_strategy == "patch":
+            pass
+        elif self.vision_feature_select_strategy == "cls_patch":
+            self.image_feature_len += 1
+        else:
+            raise ValueError(f"Unexpected select feature: {self.select_feature}")
+
+        # load mm_projector
+        projector_weights = {
+            "model.mm_projector.0": "multi_modal_projector.linear_1",
+            "model.mm_projector.2": "multi_modal_projector.linear_2",
+            "model.vision_tower.vision_tower": "vision_tower",  # Update the vision tower weights if we find them in the checkpoint (it may be finetuned).
+        }
+        params_dict = dict(self.named_parameters())
+        for name, loaded_weight in hf_model_weights_iterator(
+            model_name_or_path, cache_dir, load_format, revision
+        ):
+            # FIXME: why projector weights read two times?
+            if "projector" in name or "vision_tower" in name:
+                for weight_name, param_name in projector_weights.items():
+                    if weight_name in name:
+                        name = name.replace(weight_name, param_name)
+                param = params_dict[name]
+                weight_loader = getattr(param, "weight_loader", default_weight_loader)
+                weight_loader(param, loaded_weight)
+
+        # load language model
+        self.language_model.load_weights(
+            model_name_or_path, cache_dir, load_format, revision
+        )
+
+        monkey_path_clip_vision_embed_forward()
+
+    @property
+    def num_patches_per_side(self):
+        return self.image_size // self.patch_size
+
+
+first_call = True
+
+
+def clip_vision_embed_forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
+    batch_size = pixel_values.shape[0]
+
+    # Move this conv layer to CPU to avoid a bug in torch >= 2.1 on A10G.
+    global first_call
+    if first_call:
+        self.patch_embedding.cpu().float()
+        first_call = False
+    pixel_values = pixel_values.to(dtype=torch.float32, device="cpu")
+    patch_embeds = self.patch_embedding(pixel_values).cuda().half()
+
+    patch_embeds = patch_embeds.flatten(2).transpose(1, 2)
+
+    class_embeds = self.class_embedding.expand(batch_size, 1, -1)
+    embeddings = torch.cat([class_embeds, patch_embeds], dim=1)
+    embeddings = embeddings + self.position_embedding(self.position_ids)
+    return embeddings
+
+
+def monkey_path_clip_vision_embed_forward():
+    import transformers
+
+    setattr(
+        transformers.models.clip.modeling_clip.CLIPVisionEmbeddings,
+        "forward",
+        clip_vision_embed_forward,
+    )
+
+
+EntryClass = LlavaQwenForCausalLM

python/sglang/srt/models/qwen2.py

@@ -303,6 +303,8 @@ class Qwen2ForCausalLM(nn.Module):
                 # Skip loading extra bias for GPTQ models.
                 if name.endswith(".bias") and name not in params_dict:
                     continue
+                if name.startswith("model.vision_tower") and name not in params_dict:
+                    continue
                 param = params_dict[name]
                 weight_loader = param.weight_loader
                 weight_loader(param, loaded_weight, shard_id)
@@ -311,6 +313,8 @@ class Qwen2ForCausalLM(nn.Module):
                 # Skip loading extra bias for GPTQ models.
                 if name.endswith(".bias") and name not in params_dict:
                     continue
+                if name.startswith("model.vision_tower") and name not in params_dict:
+                    continue
                 param = params_dict[name]
                 weight_loader = getattr(param, "weight_loader", default_weight_loader)
                 weight_loader(param, loaded_weight)