Update version to v0.1.33

GitHub Actions · GitHub Actions · commit 2a7d138b2ce5 · 2025-12-18T15:01:48.000Z
diff --git a/changelog/2025-05-22.mdx b/changelog/2025-05-22.mdx
@@ -3,4 +3,4 @@ id: 2025-05-22
 date: 2025-05-22
 ---
 
-- We released Mistral OCR 2505 (`mistral-ocr-2505`) and [annotations](/capabilities/OCR/annotations). <Tag model/> <Tag api/>
+- We released Mistral OCR 2 (`mistral-ocr-2505`) and [annotations](/capabilities/OCR/annotations). <Tag model/> <Tag api/>
diff --git a/changelog/2025-12-18.mdx b/changelog/2025-12-18.mdx
@@ -0,0 +1,8 @@
+---
+id: 2025-12-18
+date: 2025-12-18
+---
+
+- We released OCR 3 (`mistral-ocr-2512`). <Tag model/>
+  - Introducing `table_format` in our OCR API, allowing you to choose between `markdown` and `html` for table formatting. <Tag api/>
+  - Introducing `extract_footer`, `extract_header` in our OCR API, as well as `hyperlinks` output. <Tag api/>
diff --git a/openapi.yaml b/openapi.yaml
diff --git a/public/img/basic_ocr_graph.png b/public/img/basic_ocr_graph.png
diff --git a/public/img/basic_ocr_graph_dark.png b/public/img/basic_ocr_graph_dark.png
diff --git a/src/app/(docs)/capabilities/document_ai/basic_ocr/ocr_with_image_base64_tab/_page.mdx b/src/app/(docs)/capabilities/document_ai/basic_ocr/ocr_with_image_base64_tab/_page.mdx
@@ -27,6 +27,7 @@ ocr_response = client.ocr.process(
         "type": "image_url",
         "image_url": f"data:image/jpeg;base64,{base64_image}" 
     },
+    # table_format="markdown",
     include_image_base64=True
 )
 ```
@@ -57,6 +58,7 @@ const ocrResponse = await client.ocr.process({
         type: "image_url",
         imageUrl: "data:image/jpeg;base64," + base64Image
     },
+    // tableFormat: "markdown",
     includeImageBase64: true
 });
 ```
@@ -74,6 +76,7 @@ curl https://api.mistral.ai/v1/ocr \
         "type": "image_url",
         "image_url": "data:image/jpeg;base64,<base64_image>"
     },
+    "table_format": "markdown",
     "include_image_base64": true
   }' -o ocr_output.json
 ```
diff --git a/src/app/(docs)/capabilities/document_ai/basic_ocr/ocr_with_image_url_tab/_page.mdx b/src/app/(docs)/capabilities/document_ai/basic_ocr/ocr_with_image_url_tab/_page.mdx
@@ -19,6 +19,7 @@ ocr_response = client.ocr.process(
         "type": "image_url",
         "image_url": "https://raw.githubusercontent.com/mistralai/cookbook/refs/heads/main/mistral/ocr/receipt.png"
     },
+    # table_format="markdown",
     include_image_base64=True
 )
 ```
@@ -39,6 +40,7 @@ const ocrResponse = await client.ocr.process({
         type: "image_url",
         imageUrl: "https://raw.githubusercontent.com/mistralai/cookbook/refs/heads/main/mistral/ocr/receipt.png",
     },
+    // tableFormat: "markdown",
     includeImageBase64: true
 });
 ```
@@ -56,6 +58,7 @@ curl https://api.mistral.ai/v1/ocr \
         "type": "image_url",
         "image_url": "https://raw.githubusercontent.com/mistralai/cookbook/refs/heads/main/mistral/ocr/receipt.png"
     },
+    "table_format": "markdown",
     "include_image_base64": true
   }' -o ocr_output.json
 ```
diff --git a/src/app/(docs)/capabilities/document_ai/basic_ocr/ocr_with_pdf_base64_tab/_page.mdx b/src/app/(docs)/capabilities/document_ai/basic_ocr/ocr_with_pdf_base64_tab/_page.mdx
diff --git a/src/app/(docs)/capabilities/document_ai/basic_ocr/ocr_with_pdf_uploaded_tab/_page.mdx b/src/app/(docs)/capabilities/document_ai/basic_ocr/ocr_with_pdf_uploaded_tab/_page.mdx
diff --git a/src/app/(docs)/capabilities/document_ai/basic_ocr/ocr_with_pdf_url_tab/_page.mdx b/src/app/(docs)/capabilities/document_ai/basic_ocr/ocr_with_pdf_url_tab/_page.mdx
diff --git a/src/app/(docs)/capabilities/document_ai/basic_ocr/page.mdx b/src/app/(docs)/capabilities/document_ai/basic_ocr/page.mdx
@@ -20,28 +20,34 @@ import OCRWithPDFUploadedTab from './ocr_with_pdf_uploaded_tab/_page.mdx';
 
 Mistral Document AI API comes with a Document OCR (Optical Character Recognition) processor, powered by our latest OCR model `mistral-ocr-latest`, which enables you to extract text and structured content from PDF documents. 
 
-<div style={{ textAlign: 'center' }}>
-  <img
-    src="/img/basic_ocr_graph.png"
-    alt="Basic OCR Graph"
-    width="600"
-    style={{ borderRadius: '15px' }}
-    className='mx-auto' 
-  />
-</div>
+<Image
+  url={['/img/basic_ocr_graph.png', '/img/basic_ocr_graph_dark.png']}
+  alt="Basic OCR Graph"
+  width="600px"
+  centered
+/>
 
 <SectionTab as="h1" sectionId="before-you-start">Before You Start</SectionTab>
 
 ### Key Features
 
-- Extracts text content while maintaining document structure and hierarchy
-- Preserves formatting like headers, paragraphs, lists and tables
-- Returns results in markdown format for easy parsing and rendering
-- Handles complex layouts including multi-column text and mixed content
+- **Extracts text** in content while maintaining document structure and hierarchy.
+- Preserves formatting like headers, paragraphs, lists and tables.
+  - **Table formatting** can be toggled between `markdown` and `html` via the `table_format` parameter.
+- Option to **extract headers and footers** via the `extract_header` and the `extract_footer` parameter, when used, the headers and footers content will be provided in the `header` and `footer` fields. By default, headers and footers are considered as part of the main content output.
+- Returns results in markdown format for easy parsing and rendering.
+- Handles complex layouts including multi-column text and mixed content and returns hyperlinks when available.
 - Processes documents at scale with high accuracy
 - Supports multiple document formats including:
     - `image_url`: png, jpeg/jpg, avif and more...
     - `document_url`: pdf, pptx, docx and more...
+    - For a non-exaustive more comprehensive list, visit our [FAQ](#faq).
+
+Learn more about our API [here](https://docs.mistral.ai/api/endpoint/ocr).
+
+:::info
+Table formatting as well as header and footer extraction is only available for OCR 2512 or newer.
+:::
 
 The OCR processor returns the extracted **text content**, **images bboxes** and metadata about the document structure, making it easy to work with the recognized content programmatically.
 
@@ -67,6 +73,36 @@ Among the PDF methods, you can use a **public available URL**, a **base64 encode
     </ExplorerTab>
 </ExplorerTabs>
 
+The output will be a JSON object containing the extracted text content, images bboxes, metadata and other information about the document structure.
+
+```py
+{
+  "pages": [ # The content of each page
+    {
+      "index": int, # The index of the corresponding page
+      "markdown": str, # The main output and raw markdown content
+      "images": list, # Image information when images are extracted
+      "tables": list, # Table information when using `table_format=html`
+      "hyperlinks": list, # Hyperlinks detected
+      "header": str|null, # Header content when using `extract_header=True`
+      "footer": str|null, # Footer content when using `extract_footer=True`
+      "dimensions": dict # The dimensions of the page
+    }
+  ],
+  "model": str, # The model used for the OCR
+  "document_annotation": dict|null, # Document annotation information when used, visit the Annotations documentation for more information
+  "usage_info": dict # Usage information
+}
+```
+
+:::note
+When extracting images, and tables using HTML, they will be replaced with placeholders, such as:
+- `![img-0.jpeg](img-0.jpeg)`
+- `[tbl-3.html](tbl-3.html)`
+
+You can map them to the actual images and tables by using the `images` and `tables` fields.
+:::
+
 <SectionTab as="h2" variant="secondary" sectionId="ocr-images">Images</SectionTab>
 
 To perform OCR on an image, you can either pass a URL to the image or directly use a Base64 encoded image.
@@ -80,6 +116,36 @@ To perform OCR on an image, you can either pass a URL to the image or directly u
     </ExplorerTab>
 </ExplorerTabs>
 
+The output will be a JSON object containing the extracted text content, images bboxes, metadata and other information about the document structure.
+
+```py
+{
+  "pages": [ # The content of each page
+    {
+      "index": int, # The index of the corresponding page
+      "markdown": str, # The main output and raw markdown content
+      "images": list, # Image information when images are extracted
+      "tables": list, # Table information when using `table_format=html`
+      "hyperlinks": list, # Hyperlinks detected
+      "header": str|null, # Header content when using `extract_header=True`
+      "footer": str|null, # Footer content when using `extract_footer=True`
+      "dimensions": dict # The dimensions of the page
+    }
+  ],
+  "model": str, # The model used for the OCR
+  "document_annotation": dict|null, # Document annotation information when used, visit the Annotations documentation for more information
+  "usage_info": dict # Usage information
+}
+```
+
+:::note
+When extracting images, and tables using HTML, they will be replaced with placeholders, such as:
+- `![img-0.jpeg](img-0.jpeg)`
+- `[tbl-3.html](tbl-3.html)`
+
+You can map them to the actual images and tables by using the `images` and `tables` fields.
+:::
+
 <SectionTab as="h1" sectionId="cookbooks">Cookbooks</SectionTab>
 
 For more information and guides on how to make use of OCR, we have the following cookbooks:
@@ -92,4 +158,24 @@ For more information and guides on how to make use of OCR, we have the following
   <FaqItem question="Are there any limits regarding the OCR API?">
     Yes, there are certain limitations for the OCR API. Uploaded document files must not exceed 50 MB in size and should be no longer than 1,000 pages.
   </FaqItem>
+  <FaqItem question="What document types are supported?">
+Our Document AI OCR Processor supports a vast range of document and image types. Below you can find a non-exhaustive list of the supported formats:
+| Documents                          | Images               |
+|------------------------------------|----------------------|
+| **PDF** (.pdf)                     | **JPEG** (.jpg, .jpeg) |
+| **Word Documents** (.docx)         | **PNG** (.png)       |
+| **PowerPoint** (.pptx)             | **AVIF** (.avif)     |
+| **Text Files** (.txt)              | **TIFF** (.tiff)     |
+| **EPUB** (.epub)                   | **GIF** (.gif)       |
+| **XML/DocBook** (.xml)             | **HEIC/HEIF** (.heic, .heif) |
+| **RTF** (.rtf)                     | **BMP** (.bmp)       |
+| **OpenDocument Text** (.odt)       | **WebP** (.webp)     |
+| **BibTeX/BibLaTeX** (.bib)         |                      |
+| **FictionBook** (.fb2)             |                      |
+| **Jupyter Notebooks** (.ipynb)     |                      |
+| **JATS XML** (.xml)                |                      |
+| **LaTeX** (.tex)                   |                      |
+| **OPML** (.opml)                   |                      |
+| **Troff** (.1, .man)               |                      |
+  </FaqItem>
 </Faq>
diff --git a/src/app/(docs)/getting-started/models/page.tsx b/src/app/(docs)/getting-started/models/page.tsx
@@ -54,7 +54,7 @@ export default function ModelsPage() {
               size="sm"
               className="font-semibold font-mono uppercase text-xs"
             >
-              <Link href="https://discord.gg/mistral" target="_blank">
+              <Link href="https://discord.gg/mistralai" target="_blank">
                 <Bullet />
                 Join our Discord ↗
               </Link>
diff --git a/src/schema/content/getting-started.ts b/src/schema/content/getting-started.ts
@@ -55,6 +55,6 @@ export const BUILDING_FEATURES = [
 
 export const HERO_AVATAR_CARDS: ModelSlug[] = [
   'ministral-3-14b-25-12',
-  'ocr-25-05',
+  'ocr-3-25-12',
   'magistral-medium-1-2-25-09',
 ] as const;
diff --git a/src/schema/content/models.ts b/src/schema/content/models.ts
@@ -7,10 +7,10 @@ export const FEATURED_MODEL_NAMES: ModelSlug[] = [
 ];
 
 export const LATEST_MODEL_NAMES: ModelSlug[] = [
+  'ocr-3-25-12',
   'mistral-small-creative-25-12',
   'devstral-2-25-12',
   'mistral-large-3-25-12',
-  'ministral-3-14b-25-12',
 ];
 
 export const FEATURED_MODEL_NAMES_MODELS_PAGE: ModelSlug[] = [
diff --git a/src/schema/models/models/index.ts b/src/schema/models/models/index.ts
@@ -2,7 +2,7 @@ import { defineModels } from '../schema';
 import mistralLarge32512 from './mistral-large-3-25-12';
 import mistralMedium312508 from './mistral-medium-3-1-25-08';
 import mistralSmall322506 from './mistral-small-3-2-25-06';
-import ocr2505 from './ocr-25-05';
+import ocr32512 from './ocr-3-25-12';
 import voxtralMiniTranscribe2507 from './voxtral-mini-transcribe-25-07';
 import codestral2508 from './codestral-25-08';
 import devstral22512 from './devstral-2-25-12';
@@ -22,6 +22,7 @@ import devstralMedium102507 from './devstral-medium-1-0-25-07';
 import devstralSmall112507 from './devstral-small-1-1-25-07';
 import magistralMedium102506 from './magistral-medium-1-0-25-06';
 import magistralSmall102506 from './magistral-small-1-0-25-06';
+import ocr22505 from './ocr-2-25-05';
 import devstralSmall102505 from './devstral-small-1-0-25-05';
 import mistralMedium32505 from './mistral-medium-3-25-05';
 import mistralSmall312503 from './mistral-small-3-1-25-03';
@@ -55,7 +56,7 @@ export const MODELS = defineModels([
   mistralLarge32512,
   mistralMedium312508,
   mistralSmall322506,
-  ocr2505,
+  ocr32512,
   voxtralMiniTranscribe2507,
   codestral2508,
   devstral22512,
@@ -75,6 +76,7 @@ export const MODELS = defineModels([
   devstralSmall112507,
   magistralMedium102506,
   magistralSmall102506,
+  ocr22505,
   devstralSmall102505,
   mistralMedium32505,
   mistralSmall312503,
diff --git a/src/schema/models/models/ministral-3b-24-1.ts b/src/schema/models/models/ministral-3b-24-1.ts
@@ -6,7 +6,7 @@ export default {
   slug: 'ministral-3b-24-1',
   releaseDate: 'October 9, 2024',
   version: '24.1',
-  frontier: true,
+  frontier: false,
   class: 'Generalist',
   type: 'Premier',
   legalButton: 'https://legal.mistral.ai/ai-governance/models/ministral-1-3b',
diff --git a/src/schema/models/models/ministral-8b-24-1.ts b/src/schema/models/models/ministral-8b-24-1.ts
@@ -6,7 +6,7 @@ export default {
   slug: 'ministral-8b-24-1',
   releaseDate: 'October 9, 2024',
   version: '24.1',
-  frontier: true,
+  frontier: false,
   class: 'Generalist',
   type: 'Premier',
   legalButton: 'https://legal.mistral.ai/ai-governance/models/ministral-1-8b',
diff --git a/src/schema/models/models/ocr-2-25-05.ts b/src/schema/models/models/ocr-2-25-05.ts
@@ -1,12 +1,12 @@
 import { StaticModel } from '../schema';
 export default {
-  name: 'OCR',
+  name: 'OCR 2',
   description: `Our OCR service powering our Document AI stack that enables our users to extract interleaved text and images`,
   shortDescription: `Our OCR service powering our Document AI stack.`,
-  slug: 'ocr-25-05',
+  slug: 'ocr-2-25-05',
   releaseDate: 'May 22, 2025',
   version: '25.05',
-  frontier: true,
+  frontier: false,
   class: 'Specialist',
   type: 'Premier',
   legalButton: 'https://legal.mistral.ai/ai-governance/models/mistral-ocr',
@@ -28,7 +28,7 @@ export default {
       
     ]
   },
-  identifiers: { apiNames: ['mistral-ocr-2505', 'mistral-ocr-latest'] },
+  identifiers: { apiNames: ['mistral-ocr-2505'] },
   capabilities: {
     input: ['image', 'document'],
     output: ['text', 'image'],
diff --git a/src/schema/models/models/ocr-3-25-12.ts b/src/schema/models/models/ocr-3-25-12.ts
@@ -0,0 +1,41 @@
+import { StaticModel } from '../schema';
+export default {
+  name: 'OCR 3',
+  description: `Our OCR service powering our Document AI stack that enables our users to extract interleaved text and images`,
+  shortDescription: `Our OCR service powering our Document AI stack.`,
+  slug: 'ocr-3-25-12',
+  releaseDate: 'December 15, 2025',
+  version: '25.12',
+  frontier: true,
+  class: 'Specialist',
+  type: 'Premier',
+  legalButton: 'https://legal.mistral.ai/ai-governance/models/mistral-ocr',
+  status: 'Active',
+  avatar: { icon: 'ocr', backgroundColor: 'red' },
+  weights: [],
+  contextLength: null,
+  ratings: { 
+    speed: { stars: 4.0, label: 'Fast' },
+    performance: { stars: 3.0, label: 'Smart' }, input: { stars: 4.0, label: 'Fast' }, output: { stars: 2.0, label: 'Moderate' } },
+  pricing: {
+    type: 'custom',
+    free: false,
+    input: [
+      { type: 'flat', price: 2.0, denominator: '/1000 Pages' },
+      { type: 'flat', price: 3.0, denominator: '/1000 Annotated Pages' }
+    ],
+    output: [
+      
+    ]
+  },
+  identifiers: { apiNames: ['mistral-ocr-2512', 'mistral-ocr-latest'] },
+  capabilities: {
+    input: ['image', 'document'],
+    output: ['text', 'image'],
+    features: ['bbox-extraction', 'ocr', 'annotations-structured-ocr', 'batching'],
+    finetuning: [],
+  },
+  metadata: {},
+  playground: '',
+  legacy: false,
+} as const satisfies StaticModel;
diff --git a/version.txt b/version.txt
@@ -1 +1,2 @@
-v0.1.33
+v0.0.66
+
