Dynamic prompt support with example

docling/datamodel/pipeline_options_vlm_model.py

@@ -1,15 +1,16 @@
 from enum import Enum
-from typing import Any, Dict, List, Literal, Optional, Union
+from typing import Any, Callable, Dict, List, Literal, Optional, Union
 
 from pydantic import AnyUrl, BaseModel
 from typing_extensions import deprecated
 
 from docling.datamodel.accelerator_options import AcceleratorDevice
+from docling.datamodel.base_models import Page
 
 
 class BaseVlmOptions(BaseModel):
     kind: str
-    prompt: str
+    prompt: Union[str, Callable[[Page], str]]
     scale: float = 2.0
     max_size: Optional[int] = None
     temperature: float = 0.0

docling/models/api_vlm_model.py

@@ -29,9 +29,6 @@ class ApiVlmModel(BasePageModel):
 
             self.timeout = self.vlm_options.timeout
             self.concurrency = self.vlm_options.concurrency
-            self.prompt_content = (
-                f"This is a page from a document.\n{self.vlm_options.prompt}"
-            )
             self.params = {
                 **self.vlm_options.params,
                 "temperature": self.vlm_options.temperature,
@@ -56,9 +53,14 @@ class ApiVlmModel(BasePageModel):
                         if hi_res_image.mode != "RGB":
                             hi_res_image = hi_res_image.convert("RGB")
 
+                    if callable(self.vlm_options.prompt):
+                        prompt = self.vlm_options.prompt(page)
+                    else:
+                        prompt = self.vlm_options.prompt
+
                     page_tags = api_image_request(
                         image=hi_res_image,
-                        prompt=self.prompt_content,
+                        prompt=prompt,
                         url=self.vlm_options.url,
                         timeout=self.timeout,
                         headers=self.vlm_options.headers,

docling/models/vlm_models_inline/hf_transformers_model.py

@@ -128,7 +128,11 @@ class HuggingFaceTransformersVlmModel(BasePageModel, HuggingFaceModelDownloadMix
                     )
 
                     # Define prompt structure
-                    prompt = self.formulate_prompt()
+                    if callable(self.vlm_options.prompt):
+                        user_prompt = self.vlm_options.prompt(page)
+                    else:
+                        user_prompt = self.vlm_options.prompt
+                    prompt = self.formulate_prompt(user_prompt)
 
                     inputs = self.processor(
                         text=prompt, images=[hi_res_image], return_tensors="pt"
@@ -162,7 +166,7 @@ class HuggingFaceTransformersVlmModel(BasePageModel, HuggingFaceModelDownloadMix
 
                 yield page
 
-    def formulate_prompt(self) -> str:
+    def formulate_prompt(self, user_prompt: str) -> str:
         """Formulate a prompt for the VLM."""
 
         if self.vlm_options.repo_id == "microsoft/Phi-4-multimodal-instruct":
@@ -173,7 +177,7 @@ class HuggingFaceTransformersVlmModel(BasePageModel, HuggingFaceModelDownloadMix
             assistant_prompt = "<|assistant|>"
             prompt_suffix = "<|end|>"
 
-            prompt = f"{user_prompt}<|image_1|>{self.vlm_options.prompt}{prompt_suffix}{assistant_prompt}"
+            prompt = f"{user_prompt}<|image_1|>{user_prompt}{prompt_suffix}{assistant_prompt}"
             _log.debug(f"prompt for {self.vlm_options.repo_id}: {prompt}")
 
             return prompt
@@ -187,7 +191,7 @@ class HuggingFaceTransformersVlmModel(BasePageModel, HuggingFaceModelDownloadMix
                         "text": "This is a page from a document.",
                     },
                     {"type": "image"},
-                    {"type": "text", "text": self.vlm_options.prompt},
+                    {"type": "text", "text": user_prompt},
                 ],
             }
         ]

docling/models/vlm_models_inline/mlx_model.py

@@ -56,8 +56,6 @@ class HuggingFaceMlxModel(BasePageModel, HuggingFaceModelDownloadMixin):
             elif (artifacts_path / repo_cache_folder).exists():
                 artifacts_path = artifacts_path / repo_cache_folder
 
-            self.param_question = vlm_options.prompt
-
             ## Load the model
             self.vlm_model, self.processor = load(artifacts_path)
             self.config = load_config(artifacts_path)
@@ -86,8 +84,12 @@ class HuggingFaceMlxModel(BasePageModel, HuggingFaceModelDownloadMixin):
                         if hi_res_image.mode != "RGB":
                             hi_res_image = hi_res_image.convert("RGB")
 
+                    if callable(self.vlm_options.prompt):
+                        user_prompt = self.vlm_options.prompt(page)
+                    else:
+                        user_prompt = self.vlm_options.prompt
                     prompt = self.apply_chat_template(
-                        self.processor, self.config, self.param_question, num_images=1
+                        self.processor, self.config, user_prompt, num_images=1
                     )
 
                     start_time = time.time()

docs/examples/vlm_pipeline_api_model.py

@@ -5,7 +5,7 @@ from pathlib import Path
 import requests
 from dotenv import load_dotenv
 
-from docling.datamodel.base_models import InputFormat
+from docling.datamodel.base_models import InputFormat, Page
 from docling.datamodel.pipeline_options import (
     VlmPipelineOptions,
 )
@@ -49,6 +49,54 @@ def ollama_vlm_options(model: str, prompt: str):
     return options
 
 
+#### Using Ollama with OlmOcr
+
+
+def ollama_olmocr_vlm_options(model: str):
+    def _dynamic_olmocr_prompt(page: Page):
+        anchor = [f"Page dimensions: {int(page.size.width)}x{int(page.size.height)}"]
+
+        for cell in page._backend.get_text_cells():
+            if not cell.text.strip():
+                continue
+            bbox = cell.to_bounding_box().to_bottom_left_origin(page.size.height)
+            anchor.append(f"[{int(bbox.l)}x{int(bbox.b)}] {cell.text}")
+
+        for rect in page._backend.get_bitmap_rects():
+            bbox = rect.to_bottom_left_origin(page.size.height)
+            anchor.append(
+                f"[Image {int(bbox.l)}x{int(bbox.b)} to {int(bbox.r)}x{int(bbox.t)}]"
+            )
+
+        if len(anchor) == 1:
+            anchor.append(
+                f"[Image 0x0 to {int(page.size.width)}x{int(page.size.height)}]"
+            )
+
+        base_text = "\n".join(anchor)
+
+        return (
+            f"Below is the image of one page of a document, as well as some raw textual"
+            f" content that was previously extracted for it. Just return the plain text"
+            f" representation of this document as if you were reading it naturally.\n"
+            f"Do not hallucinate.\n"
+            f"RAW_TEXT_START\n{base_text}\nRAW_TEXT_END"
+        )
+
+    options = ApiVlmOptions(
+        url="http://localhost:11434/v1/chat/completions",  # the default Ollama endpoint
+        params=dict(
+            model=model,
+        ),
+        prompt=_dynamic_olmocr_prompt,
+        timeout=90,
+        scale=1.0,
+        max_size=1024,  # from OlmOcr pipeline
+        response_format=ResponseFormat.MARKDOWN,
+    )
+    return options
+
+
 #### Using a cloud service like IBM watsonx.ai
 
 
@@ -130,6 +178,12 @@ def main():
     #     prompt="OCR the full page to markdown.",
     # )
 
+    # Example using the OlmOcr (dynamic prompt) model with Ollama:
+    # (uncomment the following lines)
+    # pipeline_options.vlm_options = ollama_olmocr_vlm_options(
+    #     model="hf.co/mradermacher/olmOCR-7B-0225-preview-GGUF:Q8_0",
+    # )
+
     # Another possibility is using online services, e.g. watsonx.ai.
     # Using requires setting the env variables WX_API_KEY and WX_PROJECT_ID.
     # (uncomment the following lines)