muLLM Debug Console

CPU Usage

—

— cores

RAM

—

— used of —

Swap

—

— used of —

Process PID

—

up —

Threads

—

thread count

Open FDs

—

file descriptors

Disk Read

—

since boot

Disk Write

—

since boot

GPU — loading... —°C

VRAM Used

—

— / —

Utilization

—

Power

—

Memory Clock

—

Net Sent

—

Net Recv

—

Connections

—

Last updated

—

Snapshot A

Not captured yet.

Snapshot B

Not captured yet.

Diff Analysis

No flame graph captured yet. Click Profile 10s or Quick 5s.

Thread Dump

No thread dump yet. Click Thread Dump.

(first call starts tracemalloc, second shows data)

#	File	Line	Size (KB)	Count	Share
No data. Click Sample Memory.

Avg latency:— Total cost:— Cache hit:— Local:— Queries:—

Time	Intent ID	Tier	Model	Latency	Cost	Cache
Loading...

API Key Status

Anthropic❌ Not set

OpenAI❌ Not set

Google❌ Not set

Together❌ Not set

Spend Today

$—

Last 24 hours

Server Info

Python—

FastAPI—

Platform—

Started—

Uptime—

Config (read-only)

CONCURRENCY—

ESCALATION_KEEP_ALIVE—

CACHE_THRESHOLD—

LOCAL_MODEL—

PORT8100

🔴 Emergency Shutdown

Type SHUTDOWN_MULLM below to unlock the button, then click to shut down the server process.

🩺 muLLM Self-Diagnostics

Click Run Tests or visit ?self=true to run diagnostics.