QES Research — Follow-up Characterization Paper
Data path, cache locality and energy in single-node trillion-parameter Mixture-of-Experts inference on an AMD Ryzen AI MAX+ 395 Windows PC with 128 GB unified memory.
This study follows the QES execution of the ~375 GB Kimi K2.5 model and asks what the workload reveals about high-memory unified-memory client hardware.
QES Characterization: Data-Path, Cache Locality and Energy in Single-Node Trillion-Parameter MoE Inference
The original QES white paper established the execution result. This follow-up moves from “can it run?” to “how does the workload behave while it runs?”
It documents measured data-path behaviour, locality, storage interaction and energy characteristics on the same single-node AMD platform, creating a stable baseline for continuing QES optimisation.
Execution Proof → System Characterization → Optimisation
AMD publicly demonstrated the 375 GB Kimi K2.5 workload using a four-node cluster of Ryzen AI MAX+ 395 systems, each with 128 GB of memory.
QES investigates a different deployment question:
Can that model class remain persistently available for local execution on one 128 GB system instead of requiring four coordinated machines?
QES demonstrates that it can — at substantially lower throughput than the four-node cluster, but with the workload retained on one Windows machine, leaving three equivalent systems available for other work.
QES was first demonstrated with Qwen3-235B and was then adapted and qualified for the substantially larger Kimi K2.5 architecture.
| Model | Routed MoE layers | Experts / layer | Routed-expert payload |
|---|---|---|---|
| Qwen3-235B | 94 | 128 | 137.29 GB |
| Kimi K2.5 | 60 | 384 | 367.30 GB |
Kimi represented approximately a 2.67× increase in routed-expert payload while also changing model family and expert topology.
The published claim is deliberately limited: QES has demonstrated cross-family portability across two materially different MoE architectures. It does not claim universal model agnosticism.
QES treats model integrity and evidence preservation as part of the execution architecture rather than as post-run checks.
The public work includes pre-execution verification of qualified expert stores, protection against unintended modification, continuous downstream correctness checking, evidence-preserving qualification, and fail-closed behaviour when required conditions are not satisfied.
These properties may be particularly relevant to medical, legal, defence, industrial and other controlled on-premises or high-assurance environments, where knowing which model data was executed and whether the qualified execution path remained valid may matter alongside raw inference speed.
QES does not claim regulatory, medical, legal, defence or safety certification.
The published Kimi validation used an AMD Ryzen AI MAX+ 395 with Radeon 8060S, 128 GB unified memory, Windows 11 Pro, and ordinary internal NVMe storage.
The accelerated storage path used a Samsung SSD 990 EVO Plus 2 TB authoritative expert store and a YMTC PC41Q-2TB-B verified secondary expert-store replica.
The public QES materials intentionally do not disclose reconstruction- enabling implementation details including exact scheduling ratios, expert-record formats, internal state machinery, cache/eviction implementation, memory-transfer implementation, launch environment, source paths or build recipe.