accounts/docs/bench_ollama.md
Haitao Pan 84cb83933d Add benchmarking scripts and configs:
- bench_embedding.sh / bench_ollama.sh for Ollama & embedding API tests
- hf_embedding_bench.py for HF model performance
- models.txt / models-emb.txt for test configs
- docs in bench_embedding.md / bench_ollama.md
2025-08-13 13:12:34 +08:00

466 B
Raw Blame History

1) 单场景(默认 max_tokens=200

bash docs/bench_ollama.sh

2) 指定模型清单

bash docs/bench_ollama.sh --input_config models.txt

3) 双场景(短答=64、长答=200

bash docs/bench_ollama.sh --input_config models.txt --gen-only

4) 导出 CSV双场景

CSV_OUTPUT=1 bash docs/bench_ollama.sh --input_config models.txt --gen-only

可调参数(示例)

SHORT_TOKENS=48 LONG_TOKENS=256 N=30 C=4 bash docs/bench_ollama.sh --gen-only