Live Feed/AMD/Fact Record
AMD logo
AMD
product launch 96% Confidence Gate August 29, 2026

AMD Releases ROCm 6.2 featuring Unified ROCprofiler-SDK and Native FP8 Support for MI300X GPUs

AMD released ROCm 6.2, introducing the unified ROCprofiler-SDK to replace legacy roctracer and rocprofiler profiling toolkits. The release adds native FP8 data type support (E4M3 and E5M2 formats) for AI workloads on AMD Instinct MI300X accelerators.

Verified State Diff

Comparison Mode:
- Previous State
ROCm 6.1 required developers to use legacy roctracer and rocprofiler toolkits for performance analysis and lacked native PyTorch FP8 quantization primitives on Instinct MI300X GPUs.
+ Verified New State
ROCm 6.2 provides a unified ROCprofiler-SDK alongside native FP8 (E4M3 and E5M2) support in PyTorch and vLLM for AMD Instinct MI300X accelerators.

Impact & Verification Analysis

WHO IS AFFECTED

AI software engineers, performance optimization teams, and HPC developers building or porting workloads to AMD Instinct GPUs.

WHY IT MATTERS

Consolidates GPU performance profiling tools into a single SDK while enabling FP8 precision, which is critical for doubling token throughput and reducing VRAM consumption in large language model inference.

Full Fact Overview

AMD officially released ROCm version 6.2.0 for developer tools and AI framework acceleration. The update deprecates legacy tracing and profiling libraries (roctracer and rocprofiler v1/v2) in favor of a single, standardized ROCprofiler-SDK for collecting application performance telemetry. Furthermore, ROCm 6.2 introduces native FP8 (E4M3 and E5M2) precision support within PyTorch and vLLM execution environments, enabling lower memory footprints and accelerated matrix multiplication on AMD Instinct MI300X hardware without custom kernel wrappers.

Multi-Source Evidence Chain (1)

AMD Official Changelog & Release Notesamd.com
TRACKED ENTITY
Explore all historical AMD changes
View AMD Hub ➔