Node 01Running
RTX 4090
- GPUs
- 1
- VRAM
- 24 GB
- Memory
- 64 GB
Start a new work
Private by design. Fast by default. Free to run locally.
Inference fleet
GPU devices
Active deviceNode 01
Node 02Cold standby
RTX 4090
- GPUs
- 2
- VRAM
- 48 GB
- Memory
- 64 GB
GPU candidates
| GPU model | Count |
|---|---|
| RTX A6000 | 4 |
| RTX 4090 | 4 |
| RTX 3090 | 2 |
| RTX 4070 Ti | 1 |
| RTX 3080 | 1 |
Model state
Idle
Reading
Queued
Error
Waiting for a request
Context fill
0%–
Recent requests
| Time | Status | Prompt | Reused | Output | Tok/s | VRAM hit rate | Duration |
|---|---|---|---|---|---|---|---|
| No requests yet | |||||||
Conversation cache
MCP servers
Their tools run on this PC with your rights, when the model decides to call them (in this page's chat only; switch it off in Sampling).
Model and engine
Local runtime
Connect your tools
Any OpenAI- or Anthropic-compatible client works with these addresses.
Model settings
Kept in the model's run config for every client. An empty field is the default. They take effect the next time the local model service starts.