Nodes

Operator-facing control page. Raw contracts stay grouped under Developer APIs.
Overview
JSON
Online1 Trusted0 Paused0 Policy blocked0

Node Profile

gx10-e88e / vllm / last heartbeat 1788175538
Filter table
New node: no completed work or earned credits yet.
Trust gradeC
score 50/100
Total earned-0.00
credits
Contribution30%
operator effective share
Current workno active assignment
active assignment
Execution slots0/16 active
Idle · 16 free
0 local · 0 network

Reliability

0
completed chunks/jobs
0
failed chunks/jobs
0
consecutive failures
not enough data
failure percentage
not enough data
average latency
0 ms
total latency
0
accepted results
0
rejected results
none
last success
none
last failure reason

Runtime

healthy
runtime contributed-cluster
ready
model readiness
qwen3-coder
model
local
supported modes
1788175538
last health check
contributed vllm cluster is answering at http://127.0.0.1:8000 / contribution cap does not gate a contributed cluster; detected backend is vllm
runtime notes

Runtime capacity

131072
served context tokens
not reported
runtime max sequences
9416862
KV-cache tokens
dynamic context fit
output admission

Policy

allowed
effective policy
allowed
computed policy
none
operator override
none
policy reason

Identity

node-41a616a6e5d3eb5e
node id
41a616a6e5d3eb5e
public key fingerprint
local-encrypted-fallback://mundusx/device-identity
trust path
ready
effective status
ready
last reported status

Scheduler Fit

eligible for compatible jobs; trust score 50 affects ranking
current scheduler fit

Fleet Browser

Large fleets should be controlled here with search, filters, sorting, and batched operator actions. The overview topology stays summarized so hundreds of nodes do not become visual noise.

Node
Host
Trust
Backend
State
Power
Policy
Updated
node-41a616a6e5d3eb5e
fingerprint 41a616a6e5d3eb5e
cap 30% • 70 GPU% free
gx10-e88e
signed device
unknown
local-encrypted-fallback://mundusx/device-identity
grade C · 50/100 · new
completed 0 · failed 0 · consecutive failures 0
accepted 0 · rejected 0 · last failure none
vllm
ready
reported ready
Idle · 0/16 active · 16 free
0 local · 0 network
unknown • AC • unknown
41a616a6e5d3eb5e
worker: healthy • model dir /home/a/.opengpu/models • model qwen3-coder • missing • runtime contributed-cluster • checked 1788175538 • llama-cli no • BLAS no
contributed vllm cluster is answering at http://127.0.0.1:8000 • contribution cap does not gate a contributed cluster; detected backend is vllm
allowed
1788175538
Showing 1-1 of 1 · page 1 of 1
Previous Next