[Bug] GPU Model Not Recognized - Shows "Device 1f99" Instead of "GeForce GTX 1650"

[adilborekciler]

Description of the bug

After upgrading from v6.1.0 to v6.2.0, my NVIDIA GeForce GTX 1650 GPU is no longer recognized correctly. Instead of showing the proper model name, it displays "Device 1f99" (the PCI device ID).

Environment

Dashdot Version: 6.2.0 (nvidia image)
Host OS: Fedora 42
GPU: NVIDIA GeForce GTX 1650
Driver Version: 580.95.05
Docker Setup: Docker Compose with NVIDIA Container Toolkit 1.18.0

Expected Behavior

GPU should display as:

Brand: NVIDIA
Model: GeForce GTX 1650
Memory: 4096 MiB

Actual Behavior

GPU displays as:

Brand: NVIDIA
Model: Device 1f99
Memory: 32 MiB`

API Response

json"gpu": {
  "layout": [
    {"brand":"Intel","model":"UHD Graphics","memory":256},
    {"brand":"NVIDIA","model":"Device 1f99","memory":32}
  ]
}

Verification

• nvidia-smi works correctly inside the container
• /proc/driver/nvidia/gpus/*/information shows correct model name
• Issue does NOT occur in v6.1.0 or v6.1.1
• Issue DOES occur in v6.2.0

Docker Compose Configuration
dashdot: image: mauricenino/dashdot:nvidia-6.2.0 privileged: true volumes: - /:/mnt/host:ro - /proc:/mnt/host/proc:ro environment: - NVIDIA_VISIBLE_DEVICES=all - NVIDIA_DRIVER_CAPABILITIES=all deploy: resources: reservations: devices: - driver: nvidia count: all capabilities: [gpu, utility, compute]

Additional Information
The PCI device ID 1f99 corresponds to NVIDIA GeForce GTX 1650, but it seems the GPU database mapping was broken or changed between v6.1.1 and v6.2.0.
Workaround
Downgrading to mauricenino/dashdot:nvidia-6.1.1 resolves the issue.

How to reproduce

No response

Relevant log output

Info output of dashdot cli

INFO
=========
Yarn: 4.9.2
Node: v20.19.4
Dash: 6.2.0

Cwd: /app
Hash: 4b754915784c835e5c20e6aeb1c77865d1f7d89c
Platform: Linux db883cec02c7 6.16.12-200.fc42.x86_64 #1 SMP PREEMPT_DYNAMIC Sun Oct 12 16:31:16 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Docker image: nvidia
In Docker: true
In Docker (env): true
In Podman: false

What browsers are you seeing the problem on?

No response

Where is your instance running?

No response

Additional context

No response