xserv

Go to file

Gahow Wang 9bb5c5c328 tools: add correctness + performance test scripts for Qwen3-8B

- test_correctness.py: compare prefill logits top-20 vs HF transformers
- bench_server.py: HTTP API benchmark (throughput, streaming, concurrent, EOS leak check)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-05-23 14:13:49 +08:00

crates

fix: 12 bug fixes from comprehensive review — 51 tok/s verified on RTX 5090

2026-05-23 14:13:43 +08:00

csrc

fix: 12 bug fixes from comprehensive review — 51 tok/s verified on RTX 5090

2026-05-23 14:13:43 +08:00

docs

fix: 12 bug fixes from comprehensive review — 51 tok/s verified on RTX 5090

2026-05-23 14:13:43 +08:00

tools

tools: add correctness + performance test scripts for Qwen3-8B

2026-05-23 14:13:49 +08:00

.gitignore

phase 0+1: project scaffold + xserv-cuda crate

2026-05-21 18:40:22 +08:00

Cargo.lock

fix: comprehensive review + 14 bug fixes + Phase 12/14 overhaul

2026-05-22 17:53:28 +08:00

Cargo.toml

fix: comprehensive review + 14 bug fixes + Phase 12/14 overhaul

2026-05-22 17:53:28 +08:00