https://github.com/flexflow/FlexFlow

FlexFlow Serve: Low-Latency, High-Performance LLM Serving
https://github.com/flexflow/FlexFlow

Last synced: 3 months ago
JSON representation

FlexFlow Serve: Low-Latency, High-Performance LLM Serving

• Host: GitHub
• URL: https://github.com/flexflow/FlexFlow
• Owner: flexflow
• License: apache-2.0
• Created: 2018-12-08T23:43:13.000Z (about 6 years ago)
• Default Branch: inference
• Last Pushed: 2024-10-18T23:44:35.000Z (3 months ago)
• Last Synced: 2024-10-19T01:13:16.277Z (3 months ago)
• Language: C++
• Homepage: https://flexflow.readthedocs.io
• Size: 232 MB
• Stars: 1,674
• Watchers: 32
• Forks: 224
• Open Issues: 300
• Metadata Files:
  • Readme: .github/README.md
  • Changelog: CHANGELOG.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE
  • Code of conduct: CODE_OF_CONDUCT.md

Awesome Lists containing this project

• awesome-llama-resources - FlexFlow Serve: Low-Latency, High-Performance LLM Serving - source compiler and distributed system for low latency, high performance LLM serving. (Toolkits)
• awesome-distributed-ml - FlexFlow: A Distributed Deep Learning Framework that Supports Flexible Parallelization Strategies.
• StarryDivineSky - flexflow/FlexFlow - 2.0 倍，在多节点、多 GPU 推理方面比现有系统高出 1.4-2.4 倍。 (A01_文本生成_文本对话 / 大语言对话模型及数据)
• awesome-Auto-Parallelism - FlexFlow
• awesome-AI-system - Unity: Accelerating DNN Training Through Joint Optimization of Algebraic Transformations and Parallelization OSDI'22