Inferact and AMD announce engineering collaboration for agentic inference on AMD Instinct GPUs in vLLM

AISummary

Inferact announces an engineering collaboration with AMD to strengthen production agentic inference on AMD Instinct GPUs in vLLM. AMD contributes the Instinct platform, ROCm engineering, and the AITER kernel library, while Inferact and the vLLM community handle model integration, serving features, and day-0 enablement.

The companies say each configuration will ship as a public per-model serving recipe, and that the work is already running in production.

AIWhy it matters

The post names the specific workloads and the division of work between AMD and vLLM, which clarifies what the collaboration covers for teams running agents on AMD hardware.

Post on XView on X

Inferact

@inferact

Today, we are announcing an engineering collaboration with @AMD to strengthen production agentic inference on AMD Instinct GPUs in @vllm_project.

AMD brings the Instinct platform, ROCm engineering, and the AITER kernel library, while Inferact and the vLLM community bring model integration, serving features, and day-0 enablement.

Together, we are focused on agentic workloads with long-running sessions, long prefixes, and tool calling, and each configuration will ship as a public per-model serving recipe.

This work is already running in production. It shipped with this summer’s major open-model releases, and Inferact uses it on AMD Instinct clusters for our Inferact Platform customers.

Everything this collaboration produces lands in open source. Read the full announcement: https://inferact.ai/news/inferact-collaborates-with-amd-to-expand-agentic-inference-to-vllm-on-amd-instinct-tm-gpus

Source: Inferact · x.comPublished Oct 16, 4:07 PM ET · added here 4:17 PM ET

Inferact and AMD announce engineering collaboration for agentic inference on AMD Instinct GPUs in vLLM

Topics