Configuration
- LLM Provider
- llama-cpp
- Ollama URL
- http://localhost:11434/api
- OpenAI URL
- http://localhost:8080/v1
- Model
- qwen3.5:4b
- LLM Temperature
- 0.2
- Num Ctx
- 4096
- TopK
- 40
- TopP
- 0.05
- Repeat Penalty
- 1.05
- Stream Response
- 1
- Thinking
- 0
- Show Performance Stats
- 1
- Vector DB
- qdrant
- Vector DB URL
- localhost
- Vector DB Port
- 6334
- Vector DB Collection
- my-collection
- Vector Search Results
- 6
- Min Score
- 0.62
- Vector Distance
- Cosine
- Vector Size
- 384
- HNSW EF
- 256
- HNSW M
- 32
- HNSW OnDisk
- 0
- Text Chunk Size
- 512
- Text Overlap
- 100
- Merge References
- 1
System Template
You are an expert assistant. If any information is provided below, use it for context:
-{0}
-