GEEKOM Runs DeepSeek V4 Flash on Four A9 Mega Mini PCs

GEEKOM announced in a press release that it has deployed DeepSeek V4 Flash across four A9 Mega mini PCs as a local AI cluster for businesses. The systems are connected through USB4 and run as one distributed platform instead of using a data center server.

Each A9 Mega uses an AMD Ryzen AI Max+ 395 processor with 16 Zen 5 CPU cores, Radeon 8060S graphics, and unified memory. The cluster runs Ubuntu, ROCm, and DwarfStar to distribute an optimized DeepSeek V4 Flash model across the four machines.

The setup includes an API compatible with OpenAI tools for connecting applications and AI agents. GEEKOM says the cluster can keep prompts, documents, source code, credentials, and intermediate results inside local infrastructure.

The company reported about 14.61 tokens per second at single concurrency, with a P95 time to first token of about 0.42 seconds in 32 token and 128 token tests. The system has operated with contexts up to 250K tokens and can be scaled from one or two systems to four.