AMD-AGI repositories

70 repositories

Magpie
Public
A lightweight, general-purpose framework for evaluating GPU kernel and benchmark.
rocm inference-benchmark kernel-optimization
rocm inference-benchmark kernel-optimization kernel-evaluater
Python
•
MIT License
•15•83•5•5•Updated Sep 22, 2026Sep 22, 2026
Primus
Public
A flexible and high-performance training framework designed for large-scale foundation model training on AMD GPUs
training distributed-training amd-gpu
training distributed-training amd-gpu training-at-scale
Python
•
Other
•59•128•3•26•Updated Sep 22, 2026Sep 22, 2026
Infera
Public
More token goodput from frontier models. A distributed, SLA-aware serving mesh — disaggregated prefill/decode, KV-aware routing, and cache offload, tuned to you…
rocm llm-inference
rocm llm-inference
Python
•
MIT License
•5•21•15•13•Updated Sep 22, 2026Sep 22, 2026
TraceLens
Public
Automating analysis from trace files
performance analysis profiling
performance analysis profiling
Python
•
MIT License
•26•91•97•50•Updated Sep 22, 2026Sep 22, 2026
GEAK
Public
Generating Efficient AI-Centric Kernels
python gpu triton
python gpu triton hip brain llm kernel-optimization
Python
•
Apache License 2.0
•53•180•16•25•Updated Sep 22, 2026Sep 22, 2026
Hyperloom
Public
An agentic system that auto-optimizes LLM workloads on AMD GPUs.
python agent machine-learning
python agent machine-learning skills amd gpu optimization inference triton hip
Python
•
Other
•55•196•20•52•Updated Sep 22, 2026Sep 22, 2026
Primus-Turbo
Public
A high-performance acceleration library dedicated to large-scale model training on AMD GPUs
training distributed-training amd-gpu
training distributed-training amd-gpu training-at-scale
Python
•
Other
•30•70•7•22•Updated Sep 22, 2026Sep 22, 2026
diffusion-models-inference
Public
Diffusion model inference benchmarking, profiling and optimizing
docker benchmarking inference
docker benchmarking inference diffusion-models siloai
Python
•
MIT License
•0•0•1•5•Updated Sep 22, 2026Sep 22, 2026
Primus-SaFE
Public
Primus-SaFE(Stability and Fault Endurance)
training distributed-training training-stability
training distributed-training training-stability training-at-scale training-observability
Go
•
Other
•4•58•5•3•Updated Sep 22, 2026Sep 22, 2026
PrimusClaw
Public
LLM agent orchestration on Kubernetes: autonomous coding-agent sessions in per-session sandboxes.
python agent kubernetes
python agent kubernetes golang typescript sandbox llm
TypeScript
•
Other
•0•3•2•1•Updated Sep 22, 2026Sep 22, 2026
AgentKernelArena
Public
AgentKernelArena provides an end-to-end siloed-benchmarking environment where different LLM-powered agents—such as Cursor Agent, Claude Code, Codex, SWE-agent, …
llamas gpu-kernels agent-evaluation
llamas gpu-kernels agent-evaluation
Python
•
Apache License 2.0
•14•118•8•10•Updated Sep 22, 2026Sep 22, 2026
ALTO
Public
ALTO: Advanced Low-precision Training and Optimization
pruning quantization sparsification
pruning quantization sparsification low-precision-training
Python
•
Other
•2•0•0•5•Updated Sep 22, 2026Sep 22, 2026
vllm-2026
Public
A high-throughput and memory-efficient inference and serving engine for LLMs
Python
•
Apache License 2.0
•23k•0•0•0•Updated Sep 21, 2026Sep 21, 2026
AMD-Hybrid-Models
Public
Official repo for AMD hybrid models training and inference workflow
amd attention-mechanism mamba
amd attention-mechanism mamba mla hybrid-models kv-cache large-language-models llm kv-cache-compression zebra-llama
Python
•
Apache License 2.0
•1•12•3•1•Updated Sep 11, 2026Sep 11, 2026
Apex
Public
Agents, and RL environment, for optimizing GPU kernels on AMD ROCm using LLM agents. Benchmarks LLM serving workloads end-to-end, profiles bottleneck kernels, o…
llamas gpu-kernels optimization-pipeline
llamas gpu-kernels optimization-pipeline
Python
•
MIT License
•12•76•1•1•Updated Sep 2, 2026Sep 2, 2026
Instella-MoE
Public
Instella MoE is a fully open state-of-the-art Mixture-of-Experts language model, trained end-to-end on AMD Instinct™ GPUs.
amd rocm mixture-of-experts
amd rocm mixture-of-experts llm instella
Python
•
Other
•11•78•0•0•Updated Sep 2, 2026Sep 2, 2026
hip_kernel_llm_lab
Public
hip_kernel_llm_lab A research and engineering playground for HIP kernel optimization and Large Language Model training, including continued pretraining, superv…
training rl hip
training rl hip rocm gpu-kernels llm
Python
•
Apache License 2.0
•0•5•1•0•Updated Aug 27, 2026Aug 27, 2026
AMDLongContextServing
Public
Kimi-Linear long-context FP8 benchmark tooling for vLLM on AMD MI355X (AITER MLA head-padding + patched decode/prefill kernels).
rocm long-context vllm
rocm long-context vllm kimi-linear mi355x
Python
•
Apache License 2.0
•0•0•0•0•Updated Aug 19, 2026Aug 19, 2026
pr_pundit
Public
An agent for OSS contributions
agent oss coding-agent
agent oss coding-agent
Python
•
Apache License 2.0
•0•0•0•0•Updated Aug 17, 2026Aug 17, 2026
mlperf-common
Public
mlperf-common
logging primus mlperf
logging primus mlperf
Python
•
Apache License 2.0
•1•0•0•0•Updated Aug 12, 2026Aug 12, 2026
FarSkip-Collective
Public
Training and inference implementation of FarSkip-Collective models enabling communication-computation overlap
deep-learning pytorch mixture-of-experts
deep-learning pytorch mixture-of-experts
Python
•
Other
•0•5•0•0•Updated Jul 23, 2026Jul 23, 2026
maxtext-slurm
Public
Toolkit for launching and observing MaxText training on Slurm-managed GPU clusters
training slurm jax
training slurm jax llm llm-training brain-tio brain-genai brain-framework maxtext
Shell
•
MIT License
•2•29•0•0•Updated Jul 19, 2026Jul 19, 2026
HummingbirdXT
Public
This repository presents an efficient acceleration pipeline for Diffusion Transformer (DiT) based video generation models, optimized for AMD client-grade GPUs, …
python pytorch diffusion
python pytorch diffusion dit video-generation long-video-generation
Python
•
Apache License 2.0
•0•17•0•0•Updated Jul 13, 2026Jul 13, 2026
torchtitan-amd
Public
A PyTorch native platform for training generative AI models
training generative-ai pytorch-rocm-amd
training generative-ai pytorch-rocm-amd
Python
•
BSD 3-Clause "New" or "Revised" License
•999•17•0•1•Updated Jun 30, 2026Jun 30, 2026
gpt-fast
Public
The GPT-Fast for Multimodal Models on AMD GPUs
amd inference llama
amd inference llama rocm multimodal llava multimodal-large-language-models qwen gptfast
Python
•
BSD 3-Clause "New" or "Revised" License
•0•7•0•0•Updated Jun 30, 2026Jun 30, 2026
Instella-Math
Public
Training code for Instella-Math model
training reasoning rocm
training reasoning rocm llm
Python
•
Other
•1•9•0•0•Updated Jun 24, 2026Jun 24, 2026
AMD-LLM
Public
Training code and resources for AMD-135M language models on AMD GPUs.
rocm large-language-models amd-agi
rocm large-language-models amd-agi
Python
•
Other
•13•195•0•0•Updated Jun 23, 2026Jun 23, 2026
m3d_rocm
Public
This project is an optimized version of Matrix3D. It has better compatibility with ROCm ecosystem.
generation 3d-reconstruction world-model
generation 3d-reconstruction world-model
Python
•
MIT License
•0•3•0•0•Updated Jun 19, 2026Jun 19, 2026
prime_amd
Public
Prime repository with functionality enabled on AMD
training inference pytorch
training inference pytorch brain-perfopt
Python
•
Apache License 2.0
•0•0•1•0•Updated Jun 16, 2026Jun 16, 2026
PARD
Public
PARD: Accelerating LLM Inference with Low-Cost PARallel Draft Model Adaptation (ICLR 26)
inference llm speculative-decoding
inference llm speculative-decoding
Python
•
MIT License
•4•35•4•0•Updated Jun 10, 2026Jun 10, 2026

ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.

Navigation Menu

AMD-AGI

All

70 repositories

Magpie

Primus

Infera

TraceLens

GEAK

Hyperloom

Primus-Turbo

diffusion-models-inference

Primus-SaFE

PrimusClaw

AgentKernelArena

ALTO

vllm-2026

AMD-Hybrid-Models

Apex

Instella-MoE

hip_kernel_llm_lab

AMDLongContextServing

pr_pundit

mlperf-common

FarSkip-Collective

maxtext-slurm

HummingbirdXT

torchtitan-amd

gpt-fast

Instella-Math

AMD-LLM

m3d_rocm

prime_amd

PARD

Repositories list

70 repositories