[Model] Initialize support for Deepseek-VL2 models (#11578)

Signed-off-by: Isotr0py <[email protected]>
Co-authored-by: Cyrus Leung <[email protected]>

.buildkite/test-pipeline.yaml

@@ -52,6 +52,7 @@ steps:
   - tests/worker
   - tests/standalone_tests/lazy_torch_compile.py
   commands:
+  - pip install git+https://github.com/Isotr0py/DeepSeek-VL2.git  # Used by multimoda processing test
   - python3 standalone_tests/lazy_torch_compile.py
   - pytest -v -s mq_llm_engine # MQLLMEngine
   - pytest -v -s async_engine # AsyncLLMEngine

docs/source/models/supported_models.md

@@ -610,6 +610,13 @@ See [this page](#generative-models) for more information on how to use generativ
   -
   - ✅︎
   - ✅︎
+* - `DeepseekVLV2ForCausalLM`
+  - DeepSeek-VL2
+  - T + I<sup>+</sup>
+  - `deepseek-ai/deepseek-vl2-tiny`(WIP), `deepseek-ai/deepseek-vl2-small`, `deepseek-ai/deepseek-vl2` etc. (see note)
+  -
+  - ✅︎
+  - ✅︎
 * - `FuyuForCausalLM`
   - Fuyu
   - T + I
@@ -755,8 +762,19 @@ See [this page](#generative-models) for more information on how to use generativ
 <sup>E</sup> Pre-computed embeddings can be inputted for this modality.  
 <sup>+</sup> Multiple items can be inputted per text prompt for this modality.
 
+````{note}
+The `deepseek-ai/deepseek-vl2-tiny` is not supported yet.
+
+To use `DeepSeek-VL2` series models, you need to install a fork version `deepseek_vl2` package:
+```shell
+pip install git+https://github.com/Isotr0py/DeepSeek-VL2.git
+```
+
+Besides, to run `DeepSeek-VL2` series models, you have to pass `--hf_overrides '{"architectures": ["DeepseekVLV2ForCausalLM"]}'` when running vLLM.
+````
+
 ```{note}
-To use `TIGER-Lab/Mantis-8B-siglip-llama3`, you have pass `--hf_overrides '{"architectures": ["MantisForConditionalGeneration"]}'` when running vLLM.
+To use `TIGER-Lab/Mantis-8B-siglip-llama3`, you have to pass `--hf_overrides '{"architectures": ["MantisForConditionalGeneration"]}'` when running vLLM.
 ```
 
 ```{note}

examples/offline_inference/vision_language.py

@@ -66,6 +66,23 @@ def run_chameleon(question: str, modality: str):
     return llm, prompt, stop_token_ids
 
 
+# Deepseek-VL2
+def run_deepseek_vl2(question: str, modality: str):
+    assert modality == "image"
+
+    model_name = "deepseek-ai/deepseek-vl2-small"
+
+    llm = LLM(model=model_name,
+              max_model_len=4096,
+              max_num_seqs=2,
+              disable_mm_preprocessor_cache=args.disable_mm_preprocessor_cache,
+              hf_overrides={"architectures": ["DeepseekVLV2ForCausalLM"]})
+
+    prompt = f"<|User|>: <image>\n{question}\n\n<|Assistant|>:"
+    stop_token_ids = None
+    return llm, prompt, stop_token_ids
+
+
 # Fuyu
 def run_fuyu(question: str, modality: str):
     assert modality == "image"
@@ -498,6 +515,7 @@ def run_qwen2_vl(question: str, modality: str):
     "aria": run_aria,
     "blip-2": run_blip2,
     "chameleon": run_chameleon,
+    "deepseek_vl_v2": run_deepseek_vl2,
     "fuyu": run_fuyu,
     "glm4v": run_glm4v,
     "h2ovl_chat": run_h2ovl,

examples/offline_inference/vision_language_multi_image.py

@@ -54,6 +54,28 @@ def load_aria(question, image_urls: List[str]) -> ModelRequestData:
     )
 
 
+def load_deepseek_vl2(question: str, image_urls: List[str]):
+    model_name = "deepseek-ai/deepseek-vl2-small"
+
+    llm = LLM(model=model_name,
+              max_model_len=4096,
+              max_num_seqs=2,
+              hf_overrides={"architectures": ["DeepseekVLV2ForCausalLM"]},
+              limit_mm_per_prompt={"image": len(image_urls)})
+
+    placeholder = "".join(f"image_{i}:<image>\n"
+                          for i, _ in enumerate(image_urls, start=1))
+    prompt = f"<|User|>: {placeholder}{question}\n\n<|Assistant|>:"
+
+    return ModelRequestData(
+        llm=llm,
+        prompt=prompt,
+        stop_token_ids=None,
+        image_data=[fetch_image(url) for url in image_urls],
+        chat_template=None,
+    )
+
+
 def load_h2onvl(question: str, image_urls: List[str]) -> ModelRequestData:
     model_name = "h2oai/h2ovl-mississippi-2b"
 
@@ -372,6 +394,7 @@ def load_qwen2_vl(question, image_urls: List[str]) -> ModelRequestData:
 
 model_example_map = {
     "aria": load_aria,
+    "deepseek_vl2": load_deepseek_vl2,
     "h2ovl_chat": load_h2onvl,
     "idefics3": load_idefics3,
     "internvl_chat": load_internvl,

tests/models/decoder_only/vision_language/test_models.py

@@ -188,6 +188,33 @@
         max_tokens=8,
         dtype="bfloat16",
     ),
+    "deepseek_vl_v2": VLMTestInfo(
+        models=["deepseek-ai/deepseek-vl2-small"],
+        test_type=(VLMTestType.IMAGE, VLMTestType.MULTI_IMAGE),
+        dtype="bfloat16",
+        prompt_formatter=lambda img_prompt: f"<|User|>: {img_prompt}\n\n<|Assistant|>: ", # noqa: E501
+        max_model_len=4096,
+        max_num_seqs=2,
+        single_image_prompts=IMAGE_ASSETS.prompts({
+            "stop_sign": "<image>\nWhat's the color of the stop sign and car?",
+            "cherry_blossom": "<image>\nWhat's the color of the tower?",
+        }),
+        multi_image_prompt="image_1:<image>\nimage_2:<image>\nDescribe the two images shortly.",    # noqa: E501
+        vllm_runner_kwargs={"hf_overrides": {"architectures": ["DeepseekVLV2ForCausalLM"]}},  # noqa: E501
+        image_size_factors=[(0.10, 0.15)],
+        patch_hf_runner=model_utils.deepseekvl2_patch_hf_runner,
+        postprocess_inputs=model_utils.cast_dtype_post_processor("images"),
+        hf_output_post_proc=model_utils.deepseekvl2_trunc_hf_output,
+        stop_str=["<｜end▁of▁sentence｜>", "<｜begin▁of▁sentence｜>"],  # noqa: E501
+        num_logprobs=5,
+        marks=[
+            pytest.mark.skipif(
+                not is_flash_attn_2_available(),
+                reason="Model needs flash-attn for numeric convergence.",
+            ),
+            large_gpu_mark(min_gb=48),
+        ],
+    ),
     "fuyu": VLMTestInfo(
         models=["adept/fuyu-8b"],
         test_type=VLMTestType.IMAGE,

tests/models/decoder_only/vision_language/vlm_utils/model_utils.py

@@ -183,6 +183,14 @@ def paligemma_vllm_to_hf_output(vllm_output: RunnerOutput,
 
 
 ####### Post-processors for HF outputs
+def deepseekvl2_trunc_hf_output(hf_output: RunnerOutput,
+                                model: str) -> RunnerOutput:
+    output_ids, output_str, out_logprobs = hf_output
+    if output_str.endswith("<｜end▁of▁sentence｜>"):
+        output_str = output_str.split("<｜end▁of▁sentence｜>")[0]
+    return output_ids, output_str, out_logprobs
+
+
 def minicpmv_trunc_hf_output(hf_output: RunnerOutput,
                              model: str) -> RunnerOutput:
     output_ids, output_str, out_logprobs = hf_output
@@ -261,6 +269,34 @@ def qwen_prompt_path_encoder(
 
 
 ####### Model-specific HuggingFace runner patchers
+def deepseekvl2_patch_hf_runner(hf_model: HfRunner) -> HfRunner:
+    """Patches and returns an instance of the HfRunner to use for GLM4."""
+    hf_processor = hf_model.processor
+
+    def processor(*args, text="", images=None, **kwargs):
+        if isinstance(images, Image):
+            images = [images]
+        # inputs is a custom class instead of dict or BatchFeature
+        inputs = hf_processor(
+            *args,
+            prompt=text,
+            images=images,
+            **kwargs,
+        )
+        inputs = {
+            k: inputs[k]
+            for k in inputs.keys()  # noqa
+            if k not in ("seq_lens", "sft_format")
+        }
+        inputs = BatchEncoding(data=inputs, tensor_type="pt")
+        return inputs
+
+    hf_model.processor = processor
+    hf_model.model.get_output_embeddings = lambda: \
+        hf_model.model.language.model.embed_tokens
+    return hf_model
+
+
 def glm_patch_hf_runner(hf_model: HfRunner) -> HfRunner:
     """Patches and returns an instance of the HfRunner to use for GLM4."""
     hf_processor = hf_model.processor

tests/models/registry.py

@@ -179,6 +179,8 @@ class _HfExamplesInfo:
                                     trust_remote_code=True),
     "ChatGLMForConditionalGeneration": _HfExamplesInfo("chatglm2-6b",
                                                        is_available_online=False),
+    # TODO(Isotr0py): Use deepseek-vl2-tiny for test after it's supported
+    "DeepseekVLV2ForCausalLM": _HfExamplesInfo("deepseek-ai/deepseek-vl2-small"),   # noqa: E501
     "FuyuForCausalLM": _HfExamplesInfo("adept/fuyu-8b"),
     "H2OVLChatModel": _HfExamplesInfo("h2oai/h2ovl-mississippi-800m"),
     "InternVLChatModel": _HfExamplesInfo("OpenGVLab/InternVL2-1B",

tests/models/test_initialization.py

@@ -26,6 +26,9 @@ def test_can_initialize(model_arch):
 
     # Avoid OOM
     def hf_overrides(hf_config: PretrainedConfig) -> PretrainedConfig:
+        if hf_config.model_type == "deepseek_vl_v2":
+            hf_config.update({"architectures": ["DeepseekVLV2ForCausalLM"]})
+
         if hasattr(hf_config, "text_config"):
             text_config: PretrainedConfig = hf_config.text_config
         else:

vllm/entrypoints/chat_utils.py

@@ -403,8 +403,8 @@ def _placeholder_str(self, modality: ModalityStr,
             if model_type.startswith("llava"):
                 return self._cached_token_str(self._tokenizer,
                                               hf_config.image_token_index)
-            if model_type in ("chameleon", "internvl_chat", "NVLM_D",
-                              "h2ovl_chat"):
+            if model_type in ("chameleon", "deepseek_vl_v2", "internvl_chat",
+                              "NVLM_D", "h2ovl_chat"):
                 return "<image>"
             if model_type == "mllama":
                 return "<|image|>"

vllm/model_executor/models/deepseek_v2.py

@@ -243,7 +243,11 @@ def __init__(
                                         bias=False,
                                         quant_config=quant_config,
                                         prefix=f"{prefix}.o_proj")
-        rope_scaling["rope_type"] = 'deepseek_yarn'
+        if rope_scaling:
+            rope_scaling["rope_type"] = 'deepseek_yarn'
+            self.use_normal_rope = False
+        else:
+            self.use_normal_rope = True
         self.rotary_emb = get_rope(qk_rope_head_dim,
                                    rotary_dim=qk_rope_head_dim,
                                    max_position=max_position_embeddings,
@@ -298,7 +302,18 @@ def forward(
                      self.qk_nope_head_dim + self.v_head_dim)
         k_nope, v = kv.split([self.qk_nope_head_dim, self.v_head_dim], dim=-1)
         k_pe = latent_cache[:, :, self.kv_lora_rank:]
+
+        if self.use_normal_rope:
+            seq_len = positions.size(0)
+            ori_q_pe_shape, ori_k_pe_shape = q_pe.shape, k_pe.shape
+            q_pe = q_pe.reshape(seq_len, -1)
+            k_pe = k_pe.reshape(seq_len, -1)
+
         q_pe, k_pe = self.rotary_emb(positions, q_pe, k_pe)
+
+        if self.use_normal_rope:
+            q_pe, k_pe = q_pe.view(ori_q_pe_shape), k_pe.view(ori_k_pe_shape)
+
         q[..., self.qk_nope_head_dim:] = q_pe
         k = torch.empty_like(q)
         k[..., :self.qk_nope_head_dim] = k_nope
@@ -355,6 +370,7 @@ def __init__(
             quant_config=quant_config,
             prefix=f"{prefix}.self_attn",
         )
+
         if (config.n_routed_experts is not None
                 and layer_idx >= config.first_k_dense_replace
                 and layer_idx % config.moe_layer_freq == 0):

vllm/model_executor/models/deepseek_v3.py

@@ -251,7 +251,11 @@ def __init__(
                                         bias=False,
                                         quant_config=quant_config,
                                         prefix=f"{prefix}.o_proj")
-        rope_scaling["rope_type"] = 'deepseek_yarn'
+        if rope_scaling:
+            rope_scaling["rope_type"] = 'deepseek_yarn'
+            self.use_normal_rope = False
+        else:
+            self.use_normal_rope = True
         self.rotary_emb = get_rope(qk_rope_head_dim,
                                    rotary_dim=qk_rope_head_dim,
                                    max_position=max_position_embeddings,
@@ -306,7 +310,18 @@ def forward(
                      self.qk_nope_head_dim + self.v_head_dim)
         k_nope, v = kv.split([self.qk_nope_head_dim, self.v_head_dim], dim=-1)
         k_pe = latent_cache[:, :, self.kv_lora_rank:]
+
+        if self.use_normal_rope:
+            seq_len = positions.size(0)
+            ori_q_pe_shape, ori_k_pe_shape = q_pe.shape, k_pe.shape
+            q_pe = q_pe.reshape(seq_len, -1)
+            k_pe = k_pe.reshape(seq_len, -1)
+
         q_pe, k_pe = self.rotary_emb(positions, q_pe, k_pe)
+
+        if self.use_normal_rope:
+            q_pe, k_pe = q_pe.view(ori_q_pe_shape), k_pe.view(ori_k_pe_shape)
+
         q[..., self.qk_nope_head_dim:] = q_pe
         k = torch.empty_like(q)
         k[..., :self.qk_nope_head_dim] = k_nope
@@ -583,7 +598,8 @@ def load_weights(self, weights: Iterable[Tuple[str,
                 continue
 
             # TODO(simon): support nextn predict layers
-            if self.config.num_nextn_predict_layers > 0:
+            if hasattr(self.config, "num_nextn_predict_layers"
+                       ) and self.config.num_nextn_predict_layers > 0:
                 assert self.config.num_nextn_predict_layers == 1
                 layer_idx = self.config.num_hidden_layers
                 if name.startswith(f"model.layers.{layer_idx}"):