[EDU-6432] Edge AI reference

src/content/docs/en/pages/main-menu/reference/build/edge-application/edge-functions.mdx

@@ -22,6 +22,8 @@ With Edge Functions for **Edge Application**, you can:
 - Deploy functions easily.
 - Make use of pair programming with the Chat GPT integration.
 - Preview the outcome of the function live on the preview deployment.
+- Run AI agents directly on Azion's distributed infrastructure with ultra-low latency.
+- Build complete AI ecosystems.
 
 ---
 
@@ -48,6 +50,21 @@ By using the [Azion Runtime](/en/documentation/runtime/overview/) to develop you
 
 ---
 
+## AI Framework Support
+
+Edge Functions provides support for advanced AI workflows, enabling you to:
+
+- Run AI agents directly on Azion's distributed infrastructure with ultra-low latency, building enterprise-grade solutions that operate at the edge with reliability and scale.
+- Create sophisticated AI workflows using **LangGraph** and **LangChain** integrations, including advanced retrieval capabilities that combine vector and full-text search for complex use cases.
+- Leverage native [Edge SQL](/en/documentation/products/store/edge-sql/) integration for RAG implementations and vector storage, along with MCP Servers that enable agent-to-agent collaboration using Google's Agent2Agent (A2A) protocol.
+- Access state-of-the-art models like Mistral, Florence, Qwen, and others, that follow OpenAI's API standard, all while reducing complexity with a fully integrated AI infrastructure running at the edge of the network.
+
+Azion's **Edge AI** allows you to run AI models on Azion Runtime, and can be used together with Edge Functions to create complex AI-powered applications.
+
+<LinkButton link="/en/documentation/products/ai/edge-ai/" label="Learn more about Edge AI" severity="secondary" target="_blank" />
+
+---
+
 ## JavaScript frameworks
 
 ### Next.js

src/content/docs/en/pages/main-menu/reference/edge-ai/ai-models/baai-bge-reranker-v2-m3.mdx

@@ -0,0 +1,118 @@
+---
+title: BAAI/bge-reranker-v2-m3
+description: >-
+  BAAI/bge-reranker-v2-m3 is a lightweight reranker model with strong multilingual capabilities.
+meta_tags: 'edge ai, ai models, artificial intelligence, edge computing'
+namespace: docs_edge_ai_models_baai_bge_reranker_v2_m3
+permalink: /documentation/products/ai/edge-ai/models/baai-bge-reranker-v2-m3/
+---
+
+**BAAI/bge-reranker-v2-m3** is a lightweight reranker model with strong multilingual capabilities. It's easy to deploy and offers fast inference.
+
+## Model details
+
+| Category | Details |
+|----------|---------|
+| **Model Name** | BAAI/bge-reranker-v2-m3 |
+| **Version** | Original |
+| **Model Category** | Reranker |
+| **Size** | 568M parameters |
+| **HuggingFace Model** | [BAAI/bge-reranker-v2-m3](https://huggingface.co/BAAI/bge-reranker-v2-m3) |
+| **License** | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) |
+
+## Capabilities
+
+| Feature | Details |
+|---------|--------|
+| Azion Long-term Support (LTS) | ✅ / ❌|
+| Context Length | 8192 |
+| Supports LoRA | ❌ |
+| Input data | Text |
+
+## Usage 
+
+### Rerank example
+
+This is an example of a basic rerank request using this model:
+
+```bash
+curl -X POST \
+  http://endpoint-url/rerank \
+  -H 'Content-Type: application/json' \
+  -d '{
+        "model": "BAAI/bge-reranker-v2-m3",
+        "query": "What is deep learning?",
+        "documents": [
+          "Deep learning is a subset of machine learning that uses neural networks with many layers",
+          "The weather is nice today",
+          "Deep learning enables computers to learn from large amounts of data",
+          "I like pizza"
+        ]
+}'
+```
+
+### Score example
+
+This is an example of a basic score request using this model:
+
+```bash
+curl -X POST \
+  http://endpoint-url/score \
+  -H 'Content-Type: application/json' \
+  -d '{
+        "model": "BAAI/bge-reranker-v2-m3",
+        "text_1": "What is deep learning?",
+        "text_2": [
+          "Deep learning is a subset of machine learning that uses neural networks with many layers",
+          "The weather is nice today",
+          "Deep learning enables computers to learn from large amounts of data",
+          "I like pizza"
+        ]
+      }'
+```
+
+### Running with Edge Functions:
+
+This is an example of how to run this model using Edge Functions:
+
+```ts
+const modelResponse = await Azion.AI.run("baai-bge-reranker-v2-m3", {
+  "query": "What is deep learning?",
+  "documents": [
+    "Deep learning is a subset of machine learning that uses neural networks with many layers",
+    "The weather is nice today",
+    "Deep learning enables computers to learn from large amounts of data",
+    "I like pizza"
+  ]
+})
+```
+
+## JSON schema
+
+```json
+{
+    "$schema": "http://json-schema.org/draft-07/schema#",
+    "type": "object",
+    "required": [
+        "query",
+        "documents"
+    ],
+    "properties": {
+        "query": {
+            "type": "string"
+        },
+        "documents": {
+            "type": "array",
+            "items": {
+                "type": "string"
+            }
+        },
+        "top_n": {
+            "type": "integer"
+        },
+        "max_tokens_per_doc": {
+            "type": "integer"
+        }
+    }
+}
+```

src/content/docs/en/pages/main-menu/reference/edge-ai/ai-models/e5-mistral-7b-instruct.mdx

@@ -0,0 +1,108 @@
+---
+title: E5 Mistral 7B Instruct
+description: >-
+  E5 Mistral 7B Instruct is optimized for English text embedding, multilingual processing, customization, and long input sequences.
+meta_tags: 'edge ai, ai models, artificial intelligence, edge computing, mistral'
+namespace: docs_edge_ai_models_e5_mistral_7b_instruct
+permalink: /documentation/products/ai/edge-ai/models/e5-mistral-7b-instruct/
+---
+
+The **E5 Mistral 7B Instruct** model is optimized for English text embedding tasks, with capabilities for multilingual processing, flexible customization, and handling long input sequences, making it suitable for complex natural language processing applications.
+
+## Model details
+
+| Category | Details |
+|----------|---------|
+| **Model Name** | E5 Mistral 7B Instruct |
+| **Version** | Original |
+| **Model Category** | Embedding |
+| **Size** | 7B parameters |
+| **HuggingFace Model** | [e5-mistral-7b-instruct](https://huggingface.co/intfloat/e5-mistral-7b-instruct) |
+| **OpenAI Compatible Endpoint**| [Embeddings](https://platform.openai.com/docs/api-reference/embeddings/create) |
+| **License** | MIT |
+
+## Capabilities
+
+| Feature | Details |
+|---------|--------|
+| Azion Long-term Support (LTS) | ❌ |
+| Context Length | 32728 |
+| Supports LoRA | ✅ |
+| Input Data | Text |
+
+## Usage 
+
+### Embedding
+
+This is an example of how to use this model to generate embeddings for text input:
+
+```bash
+curl http://endpoint-url/v1/embeddings \
+ -H "Content-Type: application/json" \
+ -d '{
+    "input": "The food was delicious and the waiter...",
+    "model": "intfloat/e5-mistral-7b-instruct",
+    "encoding_format": "float"
+  }'
+```
+
+### Running with Edge Functions:
+
+This is and example of how to use this model with Edge Functions:
+
+```ts
+const modelResponse = await Azion.AI.run("intfloat-e5-mistral-7b-instruct", {
+  "input": "The food was delicious and the waiter...",
+  "encoding_format": "float"
+})
+```
+
+## JSON schema
+
+```json
+{
+    "$schema": "http://json-schema.org/draft-07/schema#",
+    "type": "object",
+    "required": [
+        "input"
+    ],
+    "properties": {
+        "encoding_format": {
+            "type": "string",
+            "enum": [
+                "float",
+                "base64"
+            ]
+        },
+        "dimensions": {
+            "type": "integer"
+        },
+        "input": {
+            "oneOf": [
+                {
+                    "type": "string"
+                },
+                {
+                    "type": "array",
+                    "items": {
+                        "oneOf": [
+                            {
+                                "type": "string"
+                            },
+                            {
+                                "type": "integer"
+                            },
+                            {
+                                "type": "array",
+                                "items": {
+                                    "type": "integer"
+                                }
+                            }
+                        ]
+                    }
+                }
+            ]
+        }
+    }
+}
+```