amirgholami/ai_and_memory_wall
AI and Memory Wall
This project offers detailed data on the computational and memory requirements of leading AI models across computer vision, natural language processing, and speech. It provides insights into the number of parameters, feature sizes, and FLOPs for both inference and training, along with a breakdown of memory usage for different model components. AI researchers and engineers can use this to understand resource demands and optimize their model selection and deployment strategies.
226 stars. No commits in the last 6 months.
Use this if you need to compare the computational cost and memory footprint of various state-of-the-art AI models for research or hardware planning.
Not ideal if you are looking for ready-to-use model implementations or a tool to benchmark your own custom models.
Stars
226
Forks
26
Language
—
License
MIT
Category
Last pushed
Mar 23, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/amirgholami/ai_and_memory_wall"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
vllm-project/vllm-ascend
Community maintained hardware plugin for vLLM on Ascend
SemiAnalysisAI/InferenceX
Open Source Continuous Inference Benchmarking Qwen3.5, DeepSeek, GPTOSS - GB200 NVL72 vs MI355X...
kvcache-ai/Mooncake
Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.
uccl-project/uccl
UCCL is an efficient communication library for GPUs, covering collectives, P2P (e.g., KV cache...
sophgo/tpu-mlir
Machine learning compiler based on MLIR for Sophgo TPU.