aibench - OpenAI API Benchmarking Tool A Go CLI tool for benchmarking OpenAI-compatible API servers. Automatically discovers available models and measures comprehensive performance metrics including requests per second, tokens per second, and response latency. Recent Benchmark Results Qwen Coder Models Comparison (30-second tests) Winner: Cerebras for raw speed, Alibaba Cloud for production reliability Key Findings Cerebtras API delivers exceptional performance (91.93 tokens/sec) but has aggressive rate limiting (73% error rate) Alibaba Cloud offers the best balance for production use with...
| Stars | 14 |
| Forks | 4 |
| Language | Go |
| Category | AI 工具 |
| License | MIT |
| Quality Score | 52.186/100 |
| Last Updated | 2026-05-03 |
| Created | 2025-08-01 |
| Platforms | go |
| Est. Tokens | ~5k |
Explore other popular ai 工具 tools:
aibench is OpenAI API Benchmarking Tool. It is categorized as a AI 工具 with 14 GitHub stars.
aibench is primarily written in Go.
You can find installation instructions and usage details in the aibench GitHub repository at github.com/steipete/aibench. The project has 14 stars and 4 forks, indicating an active community.
aibench is released under the MIT license, making it free to use and modify according to the license terms.