Leave your GPUs for your models
The original SparkBotLLM 1.7B Q4 runs on CPU using the compact native runtime. The model download is about 1.05 GB.
A little local help
When you need it
Your setup assistant runs alongside your models. Ask about your Sparks, inspect the basics and work through the next step with reviewed controls.
Add your local helperRead the model card
SparkbotPowered by SparkBotLLM
LocalRECORDED EXAMPLE
What MHz are our Sparks running at?
Here’s what your Sparks reported.
HEAD SPARK1,989 MHz
WORKER SPARK1,989 MHz
A startup service requests a 2,000 MHz cap. Measured clocks can sit just below the ceiling.
Read-only inspection · no settings changed
Your hardware.
Your conversation.
Your conversation.
MEET SPARKBOTLLMA little help
Give your Spark a local assistant
A little help
Right where you need it
You don’t have to figure out every setting alone. SparkBotLLM helps inspect your hardware, work through setup issues and review the next deployment step. The original 1.7B assistant runs locally on CPU, leaving the GPUs for your models.
- Inspect Docker, GPU clocks, memory and connections.
- Troubleshoot your setup with terminal tools.
- Review commands and administrator changes before they run.
- Keep model inference on your own hardware.
SparkBotLLM
Adapted from LiquidAI LFM3 1.7B · Q4
Review changes before they run
Read-only checks inspect your system. Proposed commands and administrator changes need review. Passwords belong in approval dialogs, never in chat.
Helper or SparkChat?
Use SparkBotLLM for setup and diagnostics. Use SparkChat for conversations with the main model you deploy.
Explore SparkChat