Context missing / error

[christianxmansilla]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:)
• Please do not modify this template :) and fill in all the required fields.

1. Is this request related to a challenge you're experiencing? Tell me about your story.

Hi, i upload <100 files (total will be about 250k onpremise) but i cant configure context, we are using Dyfi on ubuntu system

Any Any idea what we are missing?

Thanks
C.

2. Additional context or comments

I configure, all other Llm providers (deepseek for reasoner, serper for web search, Cohere for rerank, etc)

[dosubot[bot]]

Hi @christianxmansilla! I'm Dosu and I’m helping the dify team.

Here are the most common reasons for context errors with file uploads in Dify and how to troubleshoot:

1. Model compatibility: Not all models support file context. For example, DeepSeek models do not support direct file reading—only models like OpenAI, Claude, or Gemini do. If you use DeepSeek, you must extract file content as text and pass it as a string context instead of as files. This is a known limitation and cause of "Invalid context structure" errors source source.

2. Explicit context mapping: In recent Dify versions, you must explicitly map context variables in the workflow. In the LLM node, use the context variable picker to select the correct upstream variable (e.g., from a knowledge retrieval node or dataset). If this mapping is missing or incorrect, you'll see placeholders or errors instead of real data source.

3. Prompt template: The prompt must reference the context block (e.g., {{ context }} or {{ files }}) for context to be included. If context is enabled but not referenced in the prompt, Dify will show a warning and context will not be passed source.

4. Files must be indexed and visible: Uploaded files must be visible in the Dify UI and indexed for retrieval. Go to the dataset/knowledge base section and check that your files are listed and indexing is complete. You can test retrieval in the KB testing section to confirm the files are accessible source.

5. Model context size: For some model providers, the "Model context size" field is required. If it's missing, you'll get a configuration error. In Dify v1.8.0, this field was missing from the UI for some models; you can work around this by manually adding the context_size parameter via API or by switching model types in the UI. This is fixed in v1.8.1+ source.

6. No hardcoded file threshold: There is no minimum or maximum file count for context configuration. What matters is that the selected variable produces the required structure (with content and, for files, proper attachment bindings) source.

If you still see errors, please provide the exact error message and confirm:

• Which model you are using in the LLM node.
• Whether your files are visible and indexed in the dataset/knowledge base UI.
• If the context variable is mapped to the correct upstream node or dataset.

This will help pinpoint the issue further.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}