From e8db2b44c9be0db4b9fe43748582bac164831e35 Mon Sep 17 00:00:00 2001 From: Rajkumar2002-Rk Date: Fri, 30 Oct 2026 01:17:34 -0500 Subject: [PATCH 1/2] docs(evaluator): note that temperature and system prompt are fixed --- apps/docs/content/docs/workflows/blocks/evaluator.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/apps/docs/content/docs/workflows/blocks/evaluator.mdx b/apps/docs/content/docs/workflows/blocks/evaluator.mdx index eb7d64b7936..dc2e9504e8f 100644 --- a/apps/docs/content/docs/workflows/blocks/evaluator.mdx +++ b/apps/docs/content/docs/workflows/blocks/evaluator.mdx @@ -35,7 +35,7 @@ The range guides the model, but Sim does not clamp every returned score to it. V **Content** is the text to score, usually ``. Structured data is formatted as text. Images and audio are not scored directly. -**Model** defaults to `claude-sonnet-5-5`. Select a supported model and evaluate its scores on representative content. **Temperature** and **System Prompt** are available in advanced fields. Sim Cloud supplies credentials for supported hosted models; other models and self-hosted deployments can require your own credentials. +**Model** defaults to `claude-sonnet-5-5`. Select a supported model and evaluate its scores on representative content. Temperature and the system prompt aren't configurable: scoring runs at temperature 0.1, and the system prompt is generated from your metrics. Put scoring guidance in the metric descriptions or the content. Sim Cloud supplies credentials for supported hosted models; other models and self-hosted deployments can require your own credentials. ### Fallback models From 0311b261843071120052d5d873e9205081b15387 Mon Sep 17 00:00:00 2001 From: Raj Kumar Nelluri <91560163+Rajkumar2002-Rk@users.noreply.github.com> Date: Fri, 20 Nov 2026 18:40:46 -0600 Subject: [PATCH 2/2] docs(evaluator): temperature only reaches models that accept one --- apps/docs/content/docs/workflows/blocks/evaluator.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/apps/docs/content/docs/workflows/blocks/evaluator.mdx b/apps/docs/content/docs/workflows/blocks/evaluator.mdx index dc2e9504e8f..80e4611c286 100644 --- a/apps/docs/content/docs/workflows/blocks/evaluator.mdx +++ b/apps/docs/content/docs/workflows/blocks/evaluator.mdx @@ -35,7 +35,7 @@ The range guides the model, but Sim does not clamp every returned score to it. V **Content** is the text to score, usually ``. Structured data is formatted as text. Images and audio are not scored directly. -**Model** defaults to `claude-sonnet-5-5`. Select a supported model and evaluate its scores on representative content. Temperature and the system prompt aren't configurable: scoring runs at temperature 0.1, and the system prompt is generated from your metrics. Put scoring guidance in the metric descriptions or the content. Sim Cloud supplies credentials for supported hosted models; other models and self-hosted deployments can require your own credentials. +**Model** defaults to `claude-sonnet-5-5`. Select a supported model and evaluate its scores on representative content. Temperature and the system prompt aren't configurable. Sim sends a temperature of 0.1 only to models that accept one, and the default model doesn't, so none is sent for it. The system prompt is generated from your metrics, so put scoring guidance in the metric descriptions. Sim Cloud supplies credentials for supported hosted models; other models and self-hosted deployments can require your own credentials. ### Fallback models