diff --git a/integrations/llms/vertex-ai.mdx b/integrations/llms/vertex-ai.mdx index 08a0c4f4..1100aec6 100644 --- a/integrations/llms/vertex-ai.mdx +++ b/integrations/llms/vertex-ai.mdx @@ -1193,6 +1193,32 @@ Grounding is invoked by passing the `google_search` tool (for newer models like If you mix regular tools with grounding tools, vertex might throw an error saying only one tool can be used at a time. +## Grounding with Google Maps + +Vertex AI also supports grounding with Google Maps for location-based queries. This allows your LLM to retrieve and use location data from Google Maps. + +To use Google Maps grounding, pass the `google_maps` (or `googleMaps`) tool in the `tools` array with optional retrieval configuration: + +```json +"tools": [ + { + "type": "function", + "function": { + "name": "google_maps", + "parameters": { + "retrievalConfig": { + // Optional: Configure retrieval settings + } + } + } + } +] +``` + + +Google Maps grounding is useful for queries involving locations, directions, places, and geographic information. + + ## gemini-2.0-flash-thinking-exp and other thinking/reasoning models `gemini-2.0-flash-thinking-exp` models return a Chain of Thought response along with the actual inference text, diff --git a/product/ai-gateway/multimodal-capabilities/image-generation.mdx b/product/ai-gateway/multimodal-capabilities/image-generation.mdx index 6544468e..896c1a54 100644 --- a/product/ai-gateway/multimodal-capabilities/image-generation.mdx +++ b/product/ai-gateway/multimodal-capabilities/image-generation.mdx @@ -129,6 +129,19 @@ curl "https://api.portkey.ai/v1/images/generations" \ [Create Image](/provider-endpoints/images/create-image) +### OpenAI gpt-image-1 Parameters + +For OpenAI's `gpt-image-1` model, additional parameters are supported: + +| Parameter | Type | Description | +|-----------|------|-------------| +| `moderation` | string | Content moderation level for generated images | +| `output_format` | string | Output format for the generated image | +| `output_compression` | number | Compression level (0-100) for the output image | +| `background` | string | Background style for the generated image | +| `partial_images` | number | Number of partial images to return (0-3) | +| `stream` | boolean | Whether to stream partial image results | + On completion, the request will get logged in the logs UI where the image can be viewed. (_Note that providers may remove the hosted image after a period of time, so some logs might only contain the url_) diff --git a/product/guardrails.mdx b/product/guardrails.mdx index 8bbef59e..89b259bc 100644 --- a/product/guardrails.mdx +++ b/product/guardrails.mdx @@ -88,7 +88,7 @@ Define a basic orchestration logic for your Guardrail here. -### There are 6 Types of Guardrail Actions +### There are 7 Types of Guardrail Actions | Action | State | Description | Impact | |:-------------- |:-------------------------------------- |:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | @@ -96,6 +96,8 @@ Define a basic orchestration logic for your Guardrail here. | **Async** | **FALSE** | **On Request** Run the Guardrail check **BEFORE** sending the request to the **LLM** **On Response** Run the Guardrail check **BEFORE** sending the response to the **user** | Will add latency to the requestUseful when your Guardrail critical and you want more orchestration over your request based on the Guardrail result | | **Deny** | **TRUE** | **On Request & Response** If any of the Guardrail checks **FAIL**, the request will be killed with a **446** status code. If all of the Guardrail checks **SUCCEED**, the request/response will be sent further with a **200** status code. | This is useful when your Guardrails are critical and upon them failing, you can not run the requestWe would advice running this action on a subset of your requests to first see the impact | | **Deny** | **FALSE** This is the **default** state | **On Request & Response** If any of the Guardrail checks **FAIL**, the request will STILL be sent, but with a **246** status code. If all of the Guardrail checks **SUCCEED**, the request/response will be sent further with a **200** status code. | This is useful when you want to log the Guardrail result but do not want it to affect your result | +| **Sequential** | **TRUE** | Run the Guardrail checks **sequentially** one after another. The next check only runs after the previous one completes. | Useful when checks have dependencies or when you want predictable ordering. Adds latency as checks run one at a time. | +| **Sequential** | **FALSE** This is the **default** state | Run the Guardrail checks **in parallel**. All checks execute simultaneously. | Faster execution when checks are independent of each other. | | **On Success** | **Send Feedback** | If **all of the** Guardrail checks **PASS**, append your custom defined feedback to the request | We recommend setting up this actionThis will help you build an "Evals dataset" of Guardrail results on your requests over time | | **On Failure** | **Send Feedback** | If **any of the** Guardrail checks **FAIL**, append your custom feedback to the request | We recommend setting up this actionThis will help you build an "Evals dataset" of Guardrail results on your requests over time | diff --git a/product/guardrails/creating-raw-guardrails-in-json.mdx b/product/guardrails/creating-raw-guardrails-in-json.mdx index faf03b76..6a26d16c 100644 --- a/product/guardrails/creating-raw-guardrails-in-json.mdx +++ b/product/guardrails/creating-raw-guardrails-in-json.mdx @@ -46,6 +46,7 @@ In this example: }], "deny": false, "async": false, + "sequential": false, "on_success": { "feedback": {"value": 1,"weight": 1} }, @@ -58,5 +59,6 @@ In this example, * `deny`: Is set to `TRUE` or `FALSE` * `async`: Is set to `TRUE` or `FALSE` +* `sequential`: Is set to `TRUE` or `FALSE` - when `TRUE`, guardrail checks run one after another instead of in parallel * `on_success`: Used to pass custom `feedback` * `on_failure`: Used to pass custom `feedback`