Nodes

Operator-facing control page. Raw contracts stay grouped under Developer APIs.
Overview
JSON
Online2 Trusted0 Paused0 Policy blocked0

Node Profile

DAVE / cuda / last heartbeat 1785497284
Filter table
Trust gradeC
score 58/100
Total earned5.00
credits
Contribution80%
operator effective share
Current workno active assignment
active assignment

Reliability

2
completed chunks/jobs
0
failed chunks/jobs
0
consecutive failures
0.0%
failure percentage
0 ms
average latency
0 ms
total latency
2
accepted results
0
rejected results
1785493201
last success
none
last failure reason

Runtime

healthy
runtime cuda
ready
model readiness
tensorblock/Qwen2.5-3B-Instruct-GGUF
model
local
supported modes
1785497284
last health check
persistent llama-server runtime is warm / CUDA low-VRAM profile selected for 4096 MB; advertise modest workloads only
runtime notes

Policy

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

Identity

node-90790410ed9ddb40
node id
90790410ed9ddb40
public key fingerprint
dpapi://mundusx/device-identity
trust path
ready
effective status
ready
last reported status

Scheduler Fit

chatcmpl-ms8sib5a-cv616mzf
scheduler score 15 / CUDA node is eligible / small context fits baseline capacity / gpu_available:20% / privacy:Public / model routing:tensorblock/Qwen2.5-3B-Instruct-GGUF tier strong (strong model deprioritized for simple request) / parallel_slots:0/1 / trust:50 completed:0 failed:0 consecutive_failures:0 / performance:no recent chunk telemetry
chatcmpl-ms8sj8nd-3z15nzde
scheduler score 15 / CUDA node is eligible / small context fits baseline capacity / gpu_available:20% / privacy:Public / model routing:tensorblock/Qwen2.5-3B-Instruct-GGUF tier strong (strong model deprioritized for simple request) / parallel_slots:0/1 / trust:54 completed:1 failed:0 consecutive_failures:0 / performance:no recent chunk telemetry

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-90790410ed9ddb40
fingerprint 90790410ed9ddb40
cap 80% • 20 GPU% free
DAVE
signed device
unknown
dpapi://mundusx/device-identity
grade C · 58/100 · neutral
completed 2 · failed 0 · consecutive failures 0
accepted 2 · rejected 0 · last failure none
cuda
ready
reported ready
unknown • AC • unknown
90790410ed9ddb40
worker: healthy • model dir C:\Users\batal\.opengpu\models • model tensorblock/Qwen2.5-3B-Instruct-GGUF • C:\Users\batal\.opengpu\models\tensorblock_qwen2_5-3b-instruct-gguf\Qwen2.5-3B-Instruct-Q3_K_M.gguf • runtime cuda • checked 1785497284 • llama-cli yes • BLAS no
persistent llama-server runtime is warm • CUDA low-VRAM profile selected for 4096 MB; advertise modest workloads only
allowed
1785497284
Showing 1-1 of 1 · page 1 of 1
Previous Next