Liangsheng Yin 078f06fbf4 [Refactor] Share chat encoding dispatch between serving and offline tools (#30623)

2026-07-09 02:45:42 -07:00

..

datasets

[bench] Add agentic-trace multi-turn dataset to bench_serving (#29215)

2026-07-06 19:45:44 -07:00

__init__.py

[Refactor] Benchmark Phase 1: extract utils and datasets from bench_serving (#19077)

2026-02-21 13:50:11 -08:00

bench_utils.py

[Benchmark] use flashinfer bench_gpu_time instead of triton do_bench (#20305)

2026-03-12 04:04:30 +00:00

endpoint.py

[misc] Share bench HTTP-client base-URL resolution with IPv6-compatible formatting (#28598)

2026-06-17 20:50:49 -07:00

offline_throughput.py

[MLX] Fix step-bounded profiling for bench tools on Apple Silicon (#29217)

2026-06-30 22:55:29 -07:00

one_batch_server.py

[Refactor] Share chat encoding dispatch between serving and offline tools (#30623)

2026-07-09 02:45:42 -07:00

one_batch.py

[refactor] Adopt get_parallel() everywhere and close out the parallel wrapper surface (#30492)

2026-07-09 02:09:39 -07:00

serving.py

[Bench] Add fixed-prompt mode and per-request spec accept length metrics (#30615)

2026-07-09 02:06:04 -07:00

utils.py

[Fix] DeepSeek-OCR-2 bench_serving: fix processor loading (#28483)

2026-06-17 15:29:37 -07:00