sophgo/tpu-mlir
Machine learning compiler based on MLIR for Sophgo TPU.
This tool helps machine learning engineers and AI solution developers convert their trained neural network models into highly efficient binary files for Sophgo TPUs. You provide a pre-trained model from frameworks like PyTorch, ONNX, TFLite, or Caffe, and it outputs an optimized `.bmodel` file that runs quickly on Sophgo hardware. It's especially useful for deploying large language models (LLMs) and computer vision models.
872 stars. Actively maintained with 19 commits in the last 30 days.
Use this if you need to deploy your trained AI models, particularly large language models or computer vision models, onto Sophgo hardware for optimized performance.
Not ideal if you are working with models that are not intended for Sophgo's specialized AI accelerators or if your models are not from the supported deep learning frameworks.
Stars
872
Forks
202
Language
C++
License
—
Category
Last pushed
Feb 12, 2026
Commits (30d)
19
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/sophgo/tpu-mlir"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
vllm-project/vllm-ascend
Community maintained hardware plugin for vLLM on Ascend
kvcache-ai/Mooncake
Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.
SemiAnalysisAI/InferenceX
Open Source Continuous Inference Benchmarking Qwen3.5, DeepSeek, GPTOSS - GB200 NVL72 vs MI355X...
uccl-project/uccl
UCCL is an efficient communication library for GPUs, covering collectives, P2P (e.g., KV cache...
BBuf/how-to-optim-algorithm-in-cuda
how to optimize some algorithm in cuda.