Running GLM-OCR, DeepSeek-OCR-2, and Dots.

DeepSeek's OCR-2 and Dots.mocr, along with Zhipu's GLM-OCR, can be unified behind an OpenAI-compatible /v1/chat/completions interface using a FastAPI normalizer, enabling zero refactoring for existing callers. The wrapper routes requests by model field to each vendor's endpoint and normalizes responses to OpenAI's chat completion schema, preserving structured JSON with bounding boxes and confidence scores for RAG and layout-aware summarization. Dots.mocr is recommended for local deployment on a single 24 GB VRAM GPU via vLLM.

Running GLM-OCR, DeepSeek-OCR-2, and Dots. DeepSeek /en/tags/deepseek/ 's OCR-2, and the newer Dots.mocr — each handles different document types better than the others, and wrapping them in an OpenAI-compatible /v1/chat/completions interface means zero refactoring for existing callers. Why bother with a unified wrapper Most OCR APIs return plain text or markdown. These three return structured JSON with bounding boxes, confidence scores, and reading order — critical when you're feeding output into an RAG /en/tags/rag/ chunker or a layout-aware summarizer. But each vendor ships its own SDK, auth scheme, and response schema. A thin FastAPI layer normalizes all of that. Architecture overview client → /v1/chat/completions OpenAI schema │ ├── router picks model by model field │ ├── glm-ocr → Zhipu HTTP endpoint │ ├── deepseek-ocr-2 → DeepSeek HTTP endpoint │ └── dots-mocr → local vLLM / TGI instance │ └── response normalizer → OpenAI choices 0 .message.content JSON string 1. Spin up the normalizer service python main.py from fastapi import FastAPI, HTTPException from pydantic import BaseModel from typing import Literal import httpx, os, json app = FastAPI class ChatRequest BaseModel : model: Literal "glm-ocr", "deepseek-ocr-2", "dots-mocr" messages: list dict max tokens: int = 4096 temperature: float = 0.0 ENDPOINTS = { "glm-ocr": os.getenv "GLM OCR URL", "https://open.bigmodel.cn/api/paas/v4/chat/completions" , "deepseek-ocr-2": os.getenv "DS OCR URL", "https://api.deepseek.com/v1/chat/completions" , "dots-mocr": os.getenv "DOTS URL", "http://localhost:8001/v1/chat/completions" , } HEADERS = { "glm-ocr": {"Authorization": f"Bearer {os.getenv 'GLM API KEY' }"}, "deepseek-ocr-2": {"Authorization": f"Bearer {os.getenv 'DEEPSEEK API KEY' }"}, "dots-mocr": {"Authorization": f"Bearer {os.getenv 'DOTS API KEY', 'local' }"}, } async def call upstream model: str, payload: dict - dict: async with httpx.AsyncClient timeout=120 as client: r = await client.post ENDPOINTS model , json=payload, headers=HEADERS model r.raise for status return r.json def normalize model: str, upstream: dict - dict: """Map each vendor's response to OpenAI shape with JSON content.""" if model == "glm-ocr": raw = upstream "choices" 0 "message" "content" elif model == "deepseek-ocr-2": raw = upstream "choices" 0 "message" "content" else: dots-mocr already returns JSON string in content raw = upstream "choices" 0 "message" "content" Ensure it's valid JSON string json.loads raw raises if malformed return { "id": upstream.get "id", "ocr-" + model , "object": "chat.completion", "choices": { "index": 0, "message": {"role": "assistant", "content": raw}, "finish reason": "stop" } , "usage": upstream.get "usage", {} } @app.post "/v1/chat/completions" async def chat req: ChatRequest : if req.model not in ENDPOINTS: raise HTTPException 400, f"Unknown model {req.model}" Extract image url from last user message user msg = next m for m in reversed req.messages if m "role" == "user" , None if not user msg or "image url" not in user msg.get "content", {} 0 : raise HTTPException 400, "Expected image url in last user message" payload = { "model": req.model, "messages": req.messages, "max tokens": req.max tokens, "temperature": req.temperature, } upstream = await call upstream req.model, payload return normalize req.model, upstream 2. Deploy Dots.mocr locally optional but recommended Dots.mocr runs well on a single 24 GB VRAM GPU via vLLM: docker run --gpus all -p 8001:8000 \ -v $PWD/models:/models \ vllm/vllm-openai:latest \ --model /models/dots-mocr \ --served-model-name dots-mocr \ --max-model-len 8192 \ --limit-mm-per-prompt image=4 Pull the model first: huggingface-cli download DOTS-OCR/DOTS-OCR-2.0 --local-dir ./models/dots-mocr 3. Client usage stays identical python from openai import OpenAI client = OpenAI base url="http://localhost:8000/v1", api key="dummy" GLM-OCR for Chinese dense tables resp = client.chat.completions.create model="glm-ocr", messages= { "role": "user", "content": {"type": "image url", "image url": {"url": "https://example.com/invoice.jpg"}} } , max tokens=4096 print resp.choices 0 .message.content JSON string with cells, bbox, confidence DeepSeek-OCR-2 for handwritten forms resp = client.chat.completions.create model="deepseek-ocr-2", messages= { "role": "user", "content": {"type": "image url", "image url": {"url": "https://example.com/handwritten.png"}} } Dots.mocr for multi-page PDFs local, no egress resp = client.chat.completions.create model="dots-mocr", messages= { "role": "user", "content": {"type": "image url", "image url": {"url": "file:///data/contract.pdf"}} } 4. Response schema you can count on All three normalize to this JSON structure inside content : json { "pages": { "page index": 0, "width": 2480, "height": 3508, "blocks": { "type": "table", "bbox": 120, 340, 2360, 1200 , "confidence": 0.96, "cells": {"row": 0, "col": 0, "text": "Item", "bbox": 130, 350, 400, 410 }, {"row": 0, "col": 1, "text": "Qty", Next The hype cycle promised mass adoption by 2024 — reality check → /en/news/6968/