// Official vendor repos
AMD ROCm
3 tracked official repos · 4 releases in the last 30 days.
01Latest releases
-
ROCm/FastFlowLM v1.0.9 Patch
This is a maintenance release. It fixes **`flm validate` on Linux**, which wrongly reported "No NPU device found." in v1.0.8.
-
ROCm/FastFlowLM v1.0.8 Patch New model
This release adds **`minicpm-v-4.7:1b`**, a 1B vision-language model with toggleable thinking and tool calling. It also **opens up the source of the Gemma 4 engine**, so you can build it yourself and swap individual oper
-
ROCm/HIPIFY therock-10.1 Major
therock release v10.1
-
ROCm/FastFlowLM v1.0.7 Patch New model
This release adds **`qwen3.8-mtp:27b`** — the largest dense model FastFlowLM has shipped to date, and the first to run **speculative decoding** on the NPU through its built-in MTP draft head. It also renames `hy-mt2:1.8b
02All repos
| Repo | Category | ★ | Latest | Licence |
|---|---|---|---|---|
| hipHIP: C++ Heterogeneous-Compute Interface for Portability | Kernels, compilers & acceleration | 4,418 | rocm-7.2.42026-05-29 | MIT |
| FastFlowLMRun LLMs on AMD Ryzen™ AI NPUs in minutes; purpose-built and deeply optimized for the AMD NPUs. | Inference & local runtime | 1,977 | v1.0.92026-10-16 | NOASSERTION (unverified) |
| HIPIFYHIPIFY: Convert CUDA to Portable C++ Code | Kernels, compilers & acceleration | 733 | therock-10.12026-10-05 | NOASSERTION (unverified) |