Model Parameters
To customize model settings for a conversation or a model:
- In Threads, click the Gear icon next to selected model
- Or, in Settings > Model Providers > Llama.cpp, click the gear icon next to a model for advanced settings
- Click the edit button next to a model to configure capabilities
Inference & Engine Parameters (Gear Icon)
These settings are available in the model settings modal:
| Parameter | Description |
|---|---|
| Context Size | Maximum prompt context length (how much text the model can consider at once). |
| GPU Layers | Number of model layers to offload to GPU. More layers = faster, but uses more VRAM. |
| Temperature | Controls response randomness. Lower = more focused, higher = more creative. |
| Top K | Top-K sampling. Limits next token selection to the K most likely. |
| Top P | Top-P (nucleus) sampling. Limits next token selection to a cumulative probability. |
| Min P | Minimum probability for token selection. |
| Repeat Last N | Number of tokens to consider for repeat penalty. |
| Repeat Penalty | Penalize repeating token sequences. |
| Presence Penalty | Penalize alpha presence (encourages new topics). |
| Frequency Penalty | Reduces word repetition. |
Model Capabilities (Edit Button)
These toggles are available when you click the edit button next to a model:
- Vision: Enable image input/output
- Tools: Enable advanced tools (web search, file ops, code)
- Embeddings: Enable embedding generation
- Web Search: Allow model to search the web
- Reasoning: Enable advanced reasoning features
Model Parameters
This setting defines and configures the model's behavior.
| Parameter | Description |
|---|---|
| Prompt Template | A structured format that guides how the model should respond. Contains placeholders and instructions that help shape the model's output in a consistent way. |
Engine Parameters
These settings parameters control how the model runs on your hardware.
| Parameter | Description |
|---|---|
| Number of GPU Layers (ngl) | - Controls how many layers of the model run on your GPU. - More layers on GPU generally means faster processing, but requires more GPU memory. |
| Context Length | - Controls how much text the model can consider at once. - Longer context allows the model to handle more input but uses more memory and runs slower. - The maximum context length varies with the model used. |
By default, Jan defaults to the minimum between 8192 and the model's maximum context length, you can adjust this based on your needs.