centersl/dify-docs
1
0
Fork 0
mirror of https://github.com/langgenius/dify-docs.git synced 2026-03-26 13:18:34 +07:00
Code Issues Packages Projects Releases Wiki Activity

Remove outdated content related to conversation history variables (#621)

Browse Source 
Co-authored-by: Riskey <riskey47@dify.ai>
This commit is contained in:
Riskey
2025-12-17 18:08:47 +08:00
committed by GitHub
parent 5d7556d5f2
commit eafe25ce25
1 changed files with 2 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
en/use-dify/nodes/llm.mdx
View File

@@ -11,7 +11,7 @@ The LLM node invokes language models to process text, images, and documents. It
</Frame>
<Info>
Configure at least one model provider in **System Settings → Model Providers** before using LLM nodes. See [how to install model plugins](/en/use-dify/workspace/plugins) for setup instructions.
Configure at least one model provider in **System Settings → Model Providers** before using LLM nodes.
</Info>
## Model Selection and Parameters
@@ -102,18 +102,12 @@ When processing images, you can control the detail level:
- **High detail** - Better accuracy for complex images but uses more tokens
- **Low detail** - Faster processing with fewer tokens for simple images
The default variable selector for vision is `sys.files` which automatically picks up files from the Start node.
The default variable selector for vision is `userinput.files` which automatically picks up files from the User Input node.
<Frame caption="File processing with multimodal LLMs">
<img src="https://assets-docs.dify.ai/2024/11/05b3d4a78038bc7afbb157078e3b2b26.png" alt="File Processing" />
</Frame>
For conversation history in completion models, insert conversation variables to maintain multi-turn context:
<Frame caption="Using conversation history variables">
<img src="https://assets-docs.dify.ai/dify-enterprise-mintlify/en/guides/workflow/node/b8642f8c6e3f562fceeefae83628fd68.png" alt="Conversation History" />
</Frame>
## Jinja2 Template Support
LLM prompts support Jinja2 templating for advanced variable handling. When you use Jinja2 mode (`edition_type: "jinja2"`), you can: