Unverified50% confidenceFactExact time
The test used uniform hardware: RTX 5090 32G, 96G memory, R7 9800X3D, along with a self-built evaluation set of 803 questions and 920 execution instances.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/18/2026
First Seen
Valid until: 11/1/2026
Sources
Related Entities
Related Claims
UnverifiedThe test platform is a local computing device equipped with 128G of VRAM and an NVIDIA RTX 5000-class GPU.79% similarUnverifiedThe tester used a standard ComfyUI workflow configured with the ref2va int8 pruned base model, 1.2MP resolution, 20 steps, on a single RTX 5090.74% similarUnverifiedThe test device is a 128G compute module (referred to in the video as the Thor T5000) claimed to have performance twice that of the NVIDIA DGX Spark.72% similarUnverifiedFor those pursuing maximum inference speed or running larger models, it is recommended to consider the RTX 4090, A6000, or newer Ada architecture cards.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1060152API
curl https://kongchang.com/api/v1/knowledge/claims/1060152MCP
get_claim(id=1060152)