From 097f606c3f70eb1f5020107b8fff0ef26356e57d Mon Sep 17 00:00:00 2001 From: Arianna Laudazzi <46651782+alaudazzi@users.noreply.github.com> Date: Wed, 16 Apr 2025 08:46:38 +0200 Subject: [PATCH 1/2] Update llm-observability.md --- .../observability/applications/llm-observability.md | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/solutions/observability/applications/llm-observability.md b/solutions/observability/applications/llm-observability.md index 1c86c6383..000ac2a55 100644 --- a/solutions/observability/applications/llm-observability.md +++ b/solutions/observability/applications/llm-observability.md @@ -15,12 +15,12 @@ Elastic’s end-to-end LLM observability is delivered through the following meth Elastic’s LLM integrations now support the most widely adopted models, including OpenAI, Azure OpenAI, and a diverse range of models hosted on Amazon Bedrock and Google Vertex AI. Depending on the LLM provider you choose, the following table shows which source you can use and which type of data -- log or metrics -- you can collect. -| **LLM Provider** | **Source** | **Metrics** | **Logs** | +| **LLM Provider** | **Metrics** | **Logs** | |--------|------------|------------| -| [Amazon Bedrock](https://www.elastic.co/guide/en/integrations/current/aws_bedrock.html)| [AWS CloudWatch Logs](https://github.com/elastic/integrations/tree/main/packages/aws_bedrock#compatibility) | ✅ | ✅ | -| [Azure OpenAI](https://www.elastic.co/guide/en/integrations/current/azure_openai.html)| [Azure Monitor and Event Hubs](https://github.com/elastic/integrations/tree/main/packages/azure_openai#azure-openai-integration) | ✅ | ✅ | -| [GCP Vertex AI](https://www.elastic.co/guide/en/integrations/current/gcp_vertexai.html) | [GCP Cloud Monitoring](https://github.com/elastic/integrations/tree/main/packages/gcp_vertexai#overview) | ✅ | 🚧 | -| [OpenAI](https://www.elastic.co/guide/en/integrations/current/openai.html) | [OpenAI Usage API](https://platform.openai.com/docs/api-reference/usage) | ✅| 🚧 | +| [Amazon Bedrock](https://www.elastic.co/guide/en/integrations/current/aws_bedrock.html)| ✅ | ✅ | +| [Azure OpenAI](https://www.elastic.co/guide/en/integrations/current/azure_openai.html)| ✅ | ✅ | +| [GCP Vertex AI](https://www.elastic.co/guide/en/integrations/current/gcp_vertexai.html) | ✅ | 🚧 | +| [OpenAI](https://www.elastic.co/guide/en/integrations/current/openai.html) | ✅| 🚧 | ## APM tracing for LLM models (via instrumentation) From 06d6cc02bb386b9e91f6267a1a879bccdcaa2fb0 Mon Sep 17 00:00:00 2001 From: Arianna Laudazzi <46651782+alaudazzi@users.noreply.github.com> Date: Wed, 16 Apr 2025 08:50:05 +0200 Subject: [PATCH 2/2] Update llm-observability.md --- solutions/observability/applications/llm-observability.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/solutions/observability/applications/llm-observability.md b/solutions/observability/applications/llm-observability.md index 000ac2a55..a8675010e 100644 --- a/solutions/observability/applications/llm-observability.md +++ b/solutions/observability/applications/llm-observability.md @@ -13,7 +13,7 @@ Elastic’s end-to-end LLM observability is delivered through the following meth ## Metrics and logs ingestion for LLM APIs (via Elastic integrations) -Elastic’s LLM integrations now support the most widely adopted models, including OpenAI, Azure OpenAI, and a diverse range of models hosted on Amazon Bedrock and Google Vertex AI. Depending on the LLM provider you choose, the following table shows which source you can use and which type of data -- log or metrics -- you can collect. +Elastic’s LLM integrations now support the most widely adopted models, including OpenAI, Azure OpenAI, and a diverse range of models hosted on Amazon Bedrock and Google Vertex AI. Depending on the LLM provider you choose, the following table shows which type of data -- log or metrics -- you can collect. | **LLM Provider** | **Metrics** | **Logs** | |--------|------------|------------|