Owner
AMD-AGI
12 indexed repositories · View on GitHub
-
Hyperloom
An agentic system that auto-optimizes LLM workloads on AMD GPUs.
Python · 193 stars
-
GEAK
Generating Efficient AI-Centric Kernels
Python · 178 stars
-
Primus
A flexible and high-performance training framework designed for large-scale foundation model training on AMD GPUs
Python · 128 stars
-
AgentKernelArena
AgentKernelArena provides an end-to-end siloed-benchmarking environment where different LLM-powered agents—such as Cursor Agent, Claude C...
Python · 118 stars
-
TraceLens
Automating analysis from trace files
Python · 91 stars
-
Magpie
A lightweight, general-purpose framework for evaluating GPU kernel and benchmark.
Python · 82 stars
-
Apex
Agents, and RL environment, for optimizing GPU kernels on AMD ROCm using LLM agents. Benchmarks LLM serving workloads end-to-end, profile...
Python · 76 stars
-
Primus-Turbo
A high-performance acceleration library dedicated to large-scale model training on AMD GPUs
Python · 70 stars
-
Primus-SaFE
Primus-SaFE(Stability and Fault Endurance)
Go · 58 stars
-
Infera
More token goodput from frontier models. A distributed, SLA-aware serving mesh — disaggregated prefill/decode, KV-aware routing, and cach...
Python · 21 stars
-
PrimusClaw
LLM agent orchestration on Kubernetes: autonomous coding-agent sessions in per-session sandboxes.
TypeScript · 3 stars
-
diffusion-models-inference
Diffusion model inference benchmarking, profiling and optimizing
Python · 0 stars
-
enhancement
AMD-AGI/diffusion-models-inference#71 · 1 assignee ·
-
fix
Difficulty 5/5 Over a week Newbie friendliness 35/100
-
feat
Difficulty 5/5 Over a week Newbie friendliness 35/100
-
fix
Difficulty 3/5 1-2 days Newbie friendliness 65/100
-
fix
Difficulty 3/5 1-2 days Newbie friendliness 74/100
-
optimization_journal.json records baseline_throughput = 0.0 for runs that measured a real baseline Open
Difficulty 3/5 1-2 days Newbie friendliness 72/100
-
Difficulty 4/5 3-5 days Newbie friendliness 58/100
-
awaiting-author-response
-
bug
AMD-AGI/diffusion-models-inference#63 · 1 comment · 1 assignee ·
-
Difficulty 3/5 1-2 days Newbie friendliness 35/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 35/100
-
Difficulty 4/5 3-5 days Newbie friendliness 45/100
-
Difficulty 5/5 Over a week Newbie friendliness 35/100
-
Difficulty 3/5 1-2 days Newbie friendliness 70/100
-
Difficulty 4/5 3-5 days Newbie friendliness 48/100
AMD-AGI/AgentKernelArena#108 ·
-
domain:inference type:feature
-
Difficulty 5/5 Over a week Newbie friendliness 35/100
AMD-AGI/PrimusClaw#48 ·
-
Difficulty 4/5 3-5 days Newbie friendliness 52/100
AMD-AGI/PrimusClaw#47 ·
-
Difficulty 4/5 3-5 days Newbie friendliness 45/100
-
domain:inference type:feature
-
Difficulty 5/5 Over a week Newbie friendliness 38/100
-
Difficulty 3/5 1-2 days Newbie friendliness 76/100
-
Difficulty 4/5 3-5 days Newbie friendliness 48/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 90/100
-
Difficulty 3/5 1-2 days Newbie friendliness 70/100
-
type:bug
-
domain:inference type:bug
-
--force-resume with a raised --target-gain closes immediately: target_reached_at survives the resume Opendomain:inference type:bug
-
Difficulty 3/5 1-2 days Newbie friendliness 72/100
-
Difficulty 4/5 3-5 days Newbie friendliness 55/100
-
Difficulty 3/5 1-2 days Newbie friendliness 58/100
-
Difficulty 4/5 3-5 days Newbie friendliness 48/100
-
Difficulty 3/5 1-2 days Newbie friendliness 68/100
-
Difficulty 4/5 3-5 days Newbie friendliness 38/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
-
Difficulty 3/5 1-2 days Newbie friendliness 74/100
Showing the newest 100
This page lists what was indexed most recently. Advanced filter has the whole inventory, narrowed by language, difficulty and how long a task takes.