diff --git a/app/_data/plugins/ai-proxy.yaml b/app/_data/plugins/ai-proxy.yaml
index 550af4a64b..d51fa26720 100644
--- a/app/_data/plugins/ai-proxy.yaml
+++ b/app/_data/plugins/ai-proxy.yaml
@@ -5,39 +5,65 @@ providers:
     chat:
       supported: true
       streaming: true
-      upstream_path: 'Use the LLM <code>chat</code> upstream path'
+      upstream_path: 'Uses the <code>Converse</code> and <code>ConverseStream</code> API'
       route_type: 'llm/v1/chat'
       model_example: '<a href="https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html">Use the model name for the specific LLM provider</a>'
       min_version: '3.8'
     completions:
       supported: true
       streaming: true
-      upstream_path: 'Use the LLM <code>completions</code> upstream path'
+      upstream_path: 'Uses the <code>Converse</code> and <code>ConverseStream</code> API'
       route_type: 'llm/v1/completions'
       model_example: '<a href="https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html">Use the model name for the specific LLM provider</a>'
       min_version: '3.8'
     embeddings:
       supported: true
       streaming: false
-      upstream_path: 'Use the LLM <code>embeddings</code> upstream path'
+      upstream_path: 'Uses the <code>InvokeModel</code> and <code>InvokeWithResponseStream</code> API'
       route_type: 'llm/v1/embeddings'
       model_example: '<a href="https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html">Use the model name for the specific LLM provider</a>'
       min_version: '3.11'
+    batches: # Native format from SDK only
+      supported: 'n/a'
+      streaming: false
+      upstream_path: 'Uses the <code>ModelInvocationJob</code> API'
+      route_type: 'llm/v1/batches'
+      model_example: 'n/a'
+      min_version: ''
+      note:
+        content: 'Batches processing for Bedrock is supported in the native format from SDK only'
+    files:
+      supported: 'n/a'
+      streaming: false
+      upstream_path: '/openai/files'
+      route_type: 'llm/v1/files'
+      model_example: 'n/a'
+      min_version: ''
+      note:
+        content: 'Bedrock does not have a dedicated Files API. File storage uses Google Cloud Storage, similar to AWS S3.'
     image:
       generations:
         supported: true
         streaming: false
-        upstream_path: 'Use the LLM <code>image/generations</code> upstream path'
+        upstream_path: 'Uses the <code>InvokeModel</code> API'
         route_type: 'image/v1/images/generations'
         model_example: '<a href="https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html">Use the model name for the specific LLM provider</a>'
         min_version: '3.11'
       edits:
         supported: true
         streaming: false
-        upstream_path: 'Use the LLM <code>image/edits</code> upstream path'
+        upstream_path: 'Uses the <code>InvokeModel</code> API'
         route_type: 'image/v1/images/edits'
         model_example: '<a href="https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html">Use the model name for the specific LLM provider</a>'
         min_version: '3.11'
+    video:
+      generations:
+        supported: true
+        streaming: false
+        upstream_path: 'Uses the <code>StartAsyncInvoke</code> API'
+        route_type: 'video/v1/videos/generations'
+        model_example: '<a href="https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html">Use the model name for the specific LLM provider</a>'
+        min_version: '3.13'
 
   - name: 'Anthropic'
     url_pattern: 'https://api.anthropic.com:443/{route_type_path}'
@@ -56,6 +82,15 @@ providers:
       route_type: 'llm/v1/completions'
       model_example: 'claude-2.1'
       min_version: '3.6'
+    batches: # Native format from SDK only
+      supported: 'n/a'
+      streaming: true
+      upstream_path: '/v1/messages/batches'
+      route_type: 'files/v1/batches'
+      model_example: 'n/a'
+      min_version: ''
+      note:
+        content: 'Batches processing for Antropic is supported in the native format from SDK only'
 
   - name: 'Azure'
     url_pattern: 'https://{azure_instance}.openai.azure.com:443/openai/deployments/{deployment_name}/{route_type_path}'
@@ -153,6 +188,25 @@ providers:
       route_type: 'realtime/v1/realtime'
       model_example: 'n/a'
       min_version: '3.11'
+    video:
+      generations:
+        supported: true
+        streaming: false
+        upstream_path: '/openai/v1/video/generations/jobs'
+        route_type: 'video/v1/videos/generations'
+        model_example: 'sora-2'
+        min_version: '3.13'
+
+  - name: 'Cerebras'
+    url_pattern: 'https://api.cerebras.ai/{route_type_path}'
+    min_version: '3.13'
+    chat:
+      supported: true
+      streaming: true
+      upstream_path: '/v1/chat/completions'
+      route_type: 'llm/v1/chat'
+      model_example: 'llama3.1-8b'
+      min_version: '3.13'
 
   - name: 'Cohere'
     url_pattern: 'https://api.cohere.com:443/{route_type_path}'
@@ -179,39 +233,106 @@ providers:
       model_example: 'embed-english-v3.0'
       min_version: '3.11'
 
+  - name: 'Dashscope'
+    url_pattern: 'https://dashscope.aliyuncs.com or https://dashscope-intl.aliyuncs.com'
+    min_version: '3.13'
+    chat:
+      supported: true
+      streaming: true
+      upstream_path: '/compatible-mode/v1/chat/completions'
+      route_type: 'llm/v1/chat'
+      model_example: 'qwen-plus'
+      min_version: '3.13'
+    embeddings:
+      supported: true
+      streaming: false
+      upstream_path: '/compatible-mode/v1/embeddings'
+      route_type: 'llm/v1/embeddings'
+      model_example: 'text-embedding-v1'
+      min_version: '3.13'
+    image:
+      generations:
+        supported: true
+        streaming: false
+        upstream_path: '/api/v1/services/aigc/multimodal-generation/generation'
+        route_type: 'image/v1/images/generations'
+        model_example: 'qwen-image-plus'
+        min_version: '3.13'
+      edits:
+        supported: true
+        streaming: false
+        upstream_path: '/api/v1/services/aigc/image2image/image-synthesis'
+        route_type: 'image/v1/images/edits'
+        model_example: 'qwen-image-plus'
+        min_version: '3.13'
+
   - name: 'Gemini'
     url_pattern: 'https://generativelanguage.googleapis.com'
     min_version: '3.8'
     chat:
       supported: true
       streaming: true
-      upstream_path: 'llm/v1/chat'
+      upstream_path: 'Uses <code>generateContent</code> API'
       route_type: 'llm/v1/chat'
       model_example: 'gemini-2.0-flash'
       min_version: '3.8'
-   
     embeddings:
       supported: true
       streaming: false
-      upstream_path: 'llm/v1/embeddings'
+      upstream_path: 'Uses <code>batchEmbedContents</code> API'
       route_type: 'llm/v1/embeddings'
       model_example: 'text-embedding-004'
       min_version: '3.11'
+    files: # Native format from SDK only
+      supported: 'n/a'
+      streaming: false
+      upstream_path: 'Uses <code>uploadFile</code> and <code>files</code> API'
+      route_type: 'llm/v1/files'
+      model_example: 'n/a'
+      min_version: ''
+      note:
+        content: 'Files processing for Gemini is supported in the native format from SDK only'
+    batches: # Native format from SDK only
+      supported: 'n/a'
+      streaming: false
+      upstream_path: 'Uses <code>batches</code> API'
+      route_type: 'llm/v1/batches'
+      model_example: 'n/a'
+      min_version: ''
+      note:
+        content: 'Batches processing for Gemini is supported in the native format from SDK only'
     image:
       generations:
         supported: true
         streaming: false
-        upstream_path: 'image/v1/images/generations'
+        upstream_path: 'Uses <code>generateContent</code> API'
         route_type: 'image/v1/images/generations'
         model_example: 'gemini-2.0-flash-preview-image-generation<sup>1</sup>'
         min_version: '3.11'
       edits:
         supported: true
         streaming: false
-        upstream_path: 'image/v1/images/edits'
+        upstream_path: 'Uses <code>generateContent</code> API'
         route_type: 'image/v1/images/edits'
         model_example: 'gemini-2.0-flash-preview-image-generation<sup>1</sup>'
         min_version: '3.11'
+    realtime: # Native format from SDK only
+      supported: true
+      streaming: true
+      upstream_path: 'Uses <code>BidiGenerateContent</code> API'
+      route_type: 'realtime/v1/realtime'
+      model_example: 'gemini-live-2.5-flash-preview-native-audio-09-2025'
+      min_version: '3.13'
+      note:
+        content: 'Realtime processing for Gemini is supported in the native format from SDK only'
+    video:
+      generations:
+        supported: true
+        streaming: false
+        upstream_path: 'Uses <code>predictLongRunning</code> API'
+        route_type: 'video/v1/videos/generations'
+        model_example: 'veo-3.1-generate-001'
+        min_version: '3.13'
 
   - name: 'Gemini Vertex'
     url_pattern: 'https://aiplatform.googleapis.com/'
@@ -219,39 +340,63 @@ providers:
     chat:
       supported: true
       streaming: true
-      upstream_path: 'llm/v1/chat'
+      upstream_path: 'Uses <code>generateContent</code> API'
       route_type: 'llm/v1/chat'
       model_example: 'gemini-2.0-flash'
       min_version: '3.8'
     completions:
       supported: true
       streaming: false
-      upstream_path: 'llm/v1/completions'
+      upstream_path: 'Uses <code>generateContent</code> API'
       route_type: 'llm/v1/completions'
       model_example: 'gemini-2.0-flash'
       min_version: '3.8'
     embeddings:
       supported: true
       streaming: false
-      upstream_path: 'llm/v1/embeddings'
+      upstream_path: 'Uses <code>generateContent</code> API'
       route_type: 'llm/v1/embeddings'
       model_example: 'text-embedding-004'
       min_version: '3.11'
+    files:
+      supported: 'n/a'
+      streaming: false
+      upstream_path: '/openai/files'
+      route_type: 'llm/v1/files'
+      model_example: 'n/a'
+      min_version: '3.11'
+      note:
+        content: 'Gemini Vertex does not have a dedicated Files API. File storage uses Google Cloud Storage, similar to AWS S3.'
+    batches:
+      supported: true
+      streaming: false
+      upstream_path: 'Uses <code>batchPredictionJobs</code> API'
+      route_type: 'llm/v1/batches'
+      model_example: 'n/a'
+      min_version: '3.13'
     image:
       generations:
         supported: true
         streaming: false
-        upstream_path: 'image/v1/images/generations'
+        upstream_path: 'Uses <code>generateContent</code> API'
         route_type: 'image/v1/images/generations'
         model_example: 'gemini-2.0-flash-preview-image-generation<sup>1</sup>'
         min_version: '3.11'
       edits:
         supported: true
         streaming: false
-        upstream_path: 'image/v1/images/edits'
+        upstream_path: 'Uses <code>generateContent</code> API'
         route_type: 'image/v1/images/edits'
         model_example: 'gemini-2.0-flash-preview-image-generation<sup>1</sup>'
         min_version: '3.11'
+    video:
+      generations:
+        supported: true
+        streaming: false
+        upstream_path: 'Uses <code>predictLongRunning</code> API'
+        route_type: 'video/v1/videos/generations'
+        model_example: 'veo-3.1-generate-001'
+        min_version: '3.13'
 
   - name: 'Hugging Face'
     url_pattern: 'https://api-inference.huggingface.co'
@@ -259,24 +404,25 @@ providers:
     chat:
       supported: true
       streaming: true
-      upstream_path: '/models/{model_provider}/{model_name}'
+      upstream_path: '/v1/chat/completions'
       route_type: 'llm/v1/chat'
       model_example: '<a href="https://huggingface.co/models?inference=warm&pipeline_tag=text-generation&sort=trending">Use the model name for the specific LLM provider</a>'
       min_version: '3.9'
-    completions:
-      supported: true
-      streaming: true
-      upstream_path: '/models/{model_provider}/{model_name}'
-      route_type: 'llm/v1/completions'
-      model_example: '<a href="https://huggingface.co/models?inference=warm&pipeline_tag=text-generation&sort=trending">Use the model name for the specific LLM provider</a>'
-      min_version: '3.9'
     embeddings:
       supported: true
       streaming: false
-      upstream_path: '/models/{model_provider}/{model_name}'
+      upstream_path: '/hf-inference/models/{model_name}/pipeline/feature-extraction'
       route_type: 'llm/v1/embeddings'
       model_example: '<a href="https://huggingface.co/models?pipeline_tag=feature-extraction">Use the embedding model name</a>'
       min_version: '3.11'
+    video:
+      generations:
+        supported: true
+        streaming: false
+        upstream_path: '/v1/videos'
+        route_type: 'video/v1/videos/generations'
+        model_example: '<a href="https://huggingface.co/models?pipeline_tag=video-generation">Use the video generation model name</a>'
+        min_version: '3.13'
 
   - name: 'Llama2'
     formats: 'supports Llama2 and Llama3 models and raw, OLLAMA, and OpenAI formats'
@@ -311,21 +457,14 @@ providers:
     chat:
       supported: true
       streaming: true
-      upstream_path: 'User-defined'
+      upstream_path: '/v1/chat/completions or user-defined'
       route_type: 'llm/v1/chat'
       model_example: 'mistral-tiny'
       min_version: '3.6'
-    completions:
-      supported: true
-      streaming: true
-      upstream_path: 'User-defined'
-      route_type: 'llm/v1/completions'
-      model_example: 'mistral-tiny'
-      min_version: '3.6'
     embeddings:
       supported: true
       streaming: false
-      upstream_path: 'User-defined'
+      upstream_path: '/v1/embeddings or user-defined'
       route_type: 'llm/v1/embeddings'
       model_example: 'mistral-embed'
       min_version: '3.11'
@@ -427,6 +566,40 @@ providers:
       route_type: 'realtime/v1/realtime'
       model_example: 'gpt-4o'
       min_version: '3.11'
+    video:
+      generations:
+        supported: true
+        streaming: false
+        upstream_path: 'Use the LLM <code>image/generations</code> upstream path'
+        route_type: 'video/v1/videos/generations'
+        model_example: 'sora-2'
+        min_version: '3.13'
+
+  - name: 'xAI'
+    url_pattern: 'https://api.x.ai/{route_type_path}'
+    min_version: '3.13'
+    chat:
+      supported: true
+      streaming: true
+      upstream_path: '/v1/chat/completions'
+      route_type: 'llm/v1/chat'
+      model_example: 'grok-4-1-fast-reasoning'
+      min_version: '3.13'
+    responses:
+      supported: true
+      streaming: false
+      upstream_path: '/v1/responses'
+      route_type: 'llm/v1/responses'
+      model_example: 'grok-4-1-fast-reasoning'
+      min_version: '3.13'
+    image:
+      generations:
+        supported: true
+        streaming: false
+        upstream_path: '/v1/images/generations'
+        route_type: 'image/v1/images/generations'
+        model_example: 'grok-2-image-1212'
+        min_version: '3.13'
 
   - name: 'xAI'
     url_pattern: 'https://api.x.ai:443/{route_type_path}'
diff --git a/app/_includes/plugins/ai-proxy/formats.md b/app/_includes/plugins/ai-proxy/formats.md
index 108d6cf400..c981c2a52c 100644
--- a/app/_includes/plugins/ai-proxy/formats.md
+++ b/app/_includes/plugins/ai-proxy/formats.md
@@ -99,29 +99,54 @@ columns:
   - title: Supported APIs
     key: apis
 rows:
-  - llm_format: "[`gemini`](./examples/gemini-native-routes/)"
+  - llm_format: "`gemini`"
     provider: Gemini
     apis: |
-      - `/generateContent`
-      - `/streamGenerateContent`
-  - llm_format: "[`bedrock`](./examples/bedrock-native-routes/)"
+      - `/v1beta/models/{model_name}:generateContent`
+      - `/v1beta/models/{model_name}:streamGenerateContent`
+      - `/v1beta/models/{model_name}:embedContent`
+      - `/v1beta/models/{model_name}:batchEmbedContent`
+      - `/v1beta/batches`
+      - `/upload/{file_id}/files`
+      - `/v1beta/files`
+  - llm_format: "`gemini`"
+    provider: Vertex
+    apis: |
+      - `/v1/projects/{project_id}/locations/{location}/models/{model_name}:generateContent`
+      - `/v1/projects/{project_id}/locations/{location}/models/{model_name}:streamGenerateContent`
+      - `/v1/projects/{project_id}/locations/{location}/models/{model_name}:embedContent`
+      - `/v1/projects/{project_id}/locations/{location}/models/{model_name}:batchEmbedContent`
+      - `/v1/projects/{project_id}/locations/{location}/models/{model_name}:predictLongRunning`
+      - `/v1/projects/{project_id}/locations/{location}/rankingConfigs/{config_name}:rank`
+      - `/v1/projects/{project_id}/locations/{location}/batchPredictionJobs`
+  - llm_format: "`bedrock`"
     provider: Bedrock
     apis: |
-      - `/converse`
-      - `/converse-stream`
-      - `/retrieveAndGenerate`
-      - `/retrieveAndGenerateStream`
-      - `/rerank`
-  - llm_format: "[`cohere`](./examples/cohere-native-routes/)"
+      - `/model/{model_name}/converse`
+      - `/model/{model_name}/converse-stream`
+      - `/model/{model_name}/invoke`
+      - `/model/{model_name}/invoke-with-response-stream`
+      - `/model/{model_name}/retrieveAndGenerate`
+      - `/model/{model_name}/retrieveAndGenerateStream`
+      - `/model/{model_name}/rerank`
+      - `/model/{model_name}/async-invoke`
+      - `/model-invocations`
+  - llm_format: "`cohere`"
     provider: Cohere
     apis: |
       - `/v1/rerank`
       - `/v2/rerank`
-  - llm_format: "[`huggingface`](./examples/hugging-face-native-routes/)"
+  - llm_format: "`huggingface`"
     provider: Hugging Face
     apis: |
       - `/generate`
       - `/generate_stream`
+  - llm_format: "`anthropic`"
+    provider: Hugging Face
+    apis: |
+      - `/v1/messages`
+      - `/v1/messages/batches`
+
 {% endtable %}
 <!-- vale on -->
 
@@ -133,7 +158,7 @@ The following sections detail the provider and statistic logging limitations.
 
 * **Anthropic**: Does not support `llm/v1/completions` or `llm/v1/embeddings`.
 * **Llama2**: Raw format lacks support for `llm/v1/embeddings`.
-* **Bedrock** and **Gemini**: Only support `auth.allow_override = false`.
+* **Gemini**: Only supports `auth.allow_override = false`.
 
 #### Statistics logging limitations
 
diff --git a/app/_includes/plugins/ai-proxy/grouped-upstreams.md b/app/_includes/plugins/ai-proxy/grouped-upstreams.md
index dad8e98412..16c043737b 100644
--- a/app/_includes/plugins/ai-proxy/grouped-upstreams.md
+++ b/app/_includes/plugins/ai-proxy/grouped-upstreams.md
@@ -63,6 +63,13 @@
     plugin=plugin %}
 {% endnavtab %}
 
+{% navtab "Cerebras" %}
+  {% include plugins/ai-proxy/tables/upstream-paths/upstream-paths.html
+    providers=providers
+    provider_name="Cerebras"
+    plugin=plugin %}
+{% endnavtab %}
+
 {% navtab "Cohere" %}
   {% include plugins/ai-proxy/tables/upstream-paths/upstream-paths.html
     providers=providers
@@ -76,7 +83,7 @@
     provider_name="Gemini"
     plugin=plugin %}
 {:.warning}
-> **[1]**: Kong AI Gateway does **not** support the [Imagen](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/imagen-4.0-generate-preview-06-06?inv=1&invt=Ab46EA&project=summit-demo-2022) model family. For image generation with Google Vertex AI, use [Gemini models](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-generation) instead.
+> **[1]**: Kong AI Gateway before 3.13 does **not** support the [Imagen](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/imagen-4.0-generate-preview-06-06?inv=1&invt=Ab46EA) model family. For image generation with Google Vertex AI, use [Gemini models](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-generation) instead.
 {% endnavtab %}
 
 {% navtab "Gemini Vertex" %}
@@ -85,7 +92,7 @@
     provider_name="Gemini Vertex"
     plugin=plugin %}
 {:.warning}
-> **[1]**: Kong AI Gateway does **not** support the [Imagen](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/imagen-4.0-generate-preview-06-06?inv=1&invt=Ab46EA&project=summit-demo-2022) model family. For image generation with Google Vertex AI, use [Gemini models](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-generation) instead.
+> **[1]**: Kong AI Gateway before 3.13 does **not** support the [Imagen](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/imagen-4.0-generate-preview-06-06?inv=1&invt=Ab46EA) model family. For image generation with Google Vertex AI, use [Gemini models](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-generation) instead.
 {% endnavtab %}
 
 {% navtab "Hugging Face" %}
@@ -109,4 +116,11 @@
     plugin=plugin %}
 {% endnavtab %}
 
+{% navtab "xAI" %}
+  {% include plugins/ai-proxy/tables/upstream-paths/upstream-paths.html
+    providers=providers
+    provider_name="xAI"
+    plugin=plugin %}
+{% endnavtab %}
+
 {% endnavtabs %}
diff --git a/app/_includes/plugins/ai-proxy/overview.md b/app/_includes/plugins/ai-proxy/overview.md
index 19d90cfb36..bdd355c472 100644
--- a/app/_includes/plugins/ai-proxy/overview.md
+++ b/app/_includes/plugins/ai-proxy/overview.md
@@ -51,7 +51,6 @@ features:
     examples: |
       * [`llm/v1/assistants`](./examples/assistants-route-type/)<br>
       * [`llm/v1/responses`](./examples/responses-route-type/)<br>
-      * [Secure GitHub MCP Server traffic using `llm/v1/responses` route type](/mcp/secure-mcp-traffic/)<br>
 
   - title: "Batch and files"
     description: Supports parallel LLM requests and file upload for long documents and structured input.
diff --git a/app/_includes/plugins/ai-proxy/tables/supported-providers-audio.html b/app/_includes/plugins/ai-proxy/tables/supported-providers-audio.html
index dc2ad912dd..e27465de38 100644
--- a/app/_includes/plugins/ai-proxy/tables/supported-providers-audio.html
+++ b/app/_includes/plugins/ai-proxy/tables/supported-providers-audio.html
@@ -10,11 +10,20 @@
   <tbody>
     {% for provider in include.providers %}
     <tr id="{{ provider.name }}-audio">
-      <td>{{ provider.name }}{% if provider.formats %} ({{ provider.formats }}){% endif %}</td>
+      <td>
+        {{ provider.name }}{% if provider.formats %} ({{ provider.formats }}){% endif %}
+        {% if provider.audio.speech.supported and provider.audio.speech.min_version %}
+          {% new_in {{ provider.audio.speech.min_version }} %}
+        {% elsif provider.audio.transcriptions.supported and provider.audio.transcriptions.min_version %}
+          {% new_in {{ provider.audio.transcriptions.min_version }} %}
+        {% elsif provider.audio.translations.supported and provider.audio.translations.min_version %}
+          {% new_in {{ provider.audio.translations.min_version }} %}
+        {% endif %}
+      </td>
       <td>{{ provider.audio.speech.supported | to_check }}</td>
       <td>{{ provider.audio.transcriptions.supported | to_check }}</td>
       <td>{{ provider.audio.translations.supported | to_check }}</td>
     </tr>
     {% endfor %}
   </tbody>
-</table>
+</table>
\ No newline at end of file
diff --git a/app/_includes/plugins/ai-proxy/tables/supported-providers-image.html b/app/_includes/plugins/ai-proxy/tables/supported-providers-image.html
index e5f25fa51a..de48daddc2 100644
--- a/app/_includes/plugins/ai-proxy/tables/supported-providers-image.html
+++ b/app/_includes/plugins/ai-proxy/tables/supported-providers-image.html
@@ -17,7 +17,16 @@
   <tbody>
     {% for provider in include.providers %}
     <tr id="{{ provider.name }}-image-realtime">
-      <td>{{ provider.name }}{% if provider.formats %} ({{ provider.formats }}){% endif %}</td>
+      <td>
+        {{ provider.name }}{% if provider.formats %} ({{ provider.formats }}){% endif %}
+        {% if provider.image.generations.supported and provider.image.generations.min_version %}
+          {% new_in {{ provider.image.generations.min_version }} %}
+        {% elsif provider.image.edits.supported and provider.image.edits.min_version %}
+          {% new_in {{ provider.image.edits.min_version }} %}
+        {% elsif show_realtime and provider.realtime.supported and provider.realtime.min_version %}
+          {% new_in {{ provider.realtime.min_version }} %}
+        {% endif %}
+      </td>
       <td>{{ provider.image.generations.supported | to_check }}</td>
       <td>{{ provider.image.edits.supported | to_check }}</td>
       {% if show_realtime %}
@@ -26,4 +35,4 @@
     </tr>
     {% endfor %}
   </tbody>
-</table>
+</table>
\ No newline at end of file
diff --git a/app/_includes/plugins/ai-proxy/tables/supported-providers-processing.html b/app/_includes/plugins/ai-proxy/tables/supported-providers-processing.html
index 045e8b82ca..f39cf73512 100644
--- a/app/_includes/plugins/ai-proxy/tables/supported-providers-processing.html
+++ b/app/_includes/plugins/ai-proxy/tables/supported-providers-processing.html
@@ -1,3 +1,21 @@
+{% assign note_counter = 0 %}
+{% assign notes = "" | split: "" %}
+{% assign features = "files,batches,assistants,responses" | split: "," %}
+
+{% comment %}First pass: collect all notes{% endcomment %}
+{% for provider in include.providers %}
+  {% for feature in features %}
+    {% assign feature_data = provider[feature] %}
+    {% if feature_data.note.content %}
+      {% assign note_counter = note_counter | plus: 1 %}
+      {% assign notes = notes | push: feature_data.note.content %}
+    {% endif %}
+  {% endfor %}
+{% endfor %}
+
+{% comment %}Reset counter for second pass{% endcomment %}
+{% assign current_note = 0 %}
+
 <table>
   <thead>
     <tr>
@@ -11,12 +29,36 @@
   <tbody>
     {% for provider in include.providers %}
     <tr id="{{ provider.name }}">
-      <td>{{ provider.name }}{% if provider.formats %} ({{ provider.formats }}){% endif %}</td>
-      <td>{{ provider.files.supported | to_check }}</td>
-      <td>{{ provider.batches.supported | to_check }}</td>
-      <td>{{ provider.assistants.supported | to_check }}</td>
-      <td>{{ provider.responses.supported | to_check }}</td>
+      <td>
+        {{ provider.name }}{% if provider.formats %} ({{ provider.formats }}){% endif %}
+        {% for feature in features %}
+          {% assign feature_data = provider[feature] %}
+          {% if feature_data.supported == true and feature_data.min_version != '' and feature_data.min_version %}
+            {% new_in {{ feature_data.min_version }} %}
+            {% break %}
+          {% endif %}
+        {% endfor %}
+      </td>
+      {% for feature in features %}
+        {% assign feature_data = provider[feature] %}
+        <td>
+          {% if feature_data.supported == 'n/a' %}
+            n/a{% if feature_data.note.content %}<sup>{{ current_note | plus: 1 }}</sup>{% assign current_note = current_note | plus: 1 %}{% endif %}
+          {% elsif feature_data.note.content %}
+            {{ feature_data.supported | to_check }}<sup>{{ current_note | plus: 1 }}</sup>{% assign current_note = current_note | plus: 1 %}
+          {% else %}
+            {{ feature_data.supported | to_check }}
+          {% endif %}
+        </td>
+      {% endfor %}
     </tr>
     {% endfor %}
   </tbody>
 </table>
+
+{% comment %}Render collected notes{% endcomment %}
+{% if notes.size > 0 %}
+{% for note in notes %}
+<p><sup>{{ forloop.index }})</sup> {{ note }}</p>
+{% endfor %}
+{% endif %}
\ No newline at end of file
diff --git a/app/_includes/plugins/ai-proxy/tables/supported-providers-text.html b/app/_includes/plugins/ai-proxy/tables/supported-providers-text.html
index fc08cf40fb..96c3452dc3 100644
--- a/app/_includes/plugins/ai-proxy/tables/supported-providers-text.html
+++ b/app/_includes/plugins/ai-proxy/tables/supported-providers-text.html
@@ -10,11 +10,18 @@
     <tbody>
       {% for provider in include.providers %}
       <tr id="{{ provider.name }}">
-        <td>{{ provider.name }}{% if provider.formats %} ({{ provider.formats }}){% endif %}</td>
+        <td>
+          {{ provider.name }}{% if provider.formats %} ({{ provider.formats }}){% endif %}
+          {% if provider.chat.supported and provider.chat.min_version %}
+            {% new_in {{ provider.chat.min_version }} %}
+          {% elsif provider.embeddings.supported and provider.embeddings.min_version %}
+            {% new_in {{ provider.embeddings.min_version }} %}
+          {% endif %}
+        </td>
         <td>{{ provider.chat.supported | to_check }}</td>
         <td>{{ provider.chat.streaming | to_check }}</td>
         <td>{{ provider.embeddings.supported | to_check }}</td>
       </tr>
       {% endfor %}
     </tbody>
-  </table>
+  </table>
\ No newline at end of file