Files

T

Xiaoyu ZhangandClaude Fable 5 874fc07d9b [Kernel] Migrate scattered quantization kernels to sglang.kernels (RFC #29630 , Phase 2.5, 1/7) (#30784 )

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

2026-07-13 16:17:01 +08:00

4-gpu-models

[misc] Make NaN-logit sanitization opt-in (default off) (#28829 )

2026-06-21 14:31:35 -07:00

8-gpu-models

[Fix] Fix DSA indexer fusion for NeoX RoPE (#30111 )

2026-07-04 03:20:55 -07:00

amd

[AMD ]Feat/dsv4 ep tbo prefill (#29362 )

2026-07-05 21:18:34 -07:00

ascend

Update test repository case scripts to the main community (#29939 )

2026-07-12 11:36:30 +08:00

attention

Fix hybrid attention graph hook test fixture (#30790 )

2026-07-10 23:37:04 +05:30

backends

[misc] Use --cuda-graph-max-bs-decode in tests, examples, and docs (#29591 )

2026-06-28 18:38:28 -07:00

bench_fn

bench: support random image resolutions (#30879 )

2026-07-12 08:28:56 +08:00

chunked_prefill

Localize cur_batch field in Scheduler to avoid field-based state access (#29407 )

2026-07-10 08:55:06 +08:00

constrained_decoding

[CI] Rename basic CI stage-a/b/c -> base-a/b/c for symmetry with extra CI (#25420 )

2026-05-15 18:26:55 -07:00

core

[Spec] Add DSpark: confidence-scheduled speculative decoding (#30261 )

2026-07-12 17:25:26 -05:00

Enable breakable prefill CUDA graph for DP attention (#30898 )

2026-07-12 17:10:04 -07:00

cpu

Support priority request header override (#30811 )

2026-07-10 14:36:43 -07:00

cuda_graph

Fix diffusion BCG lifetime and add Z-Image-Turbo CI (#30584 )

2026-07-10 21:19:59 +08:00

dcp

[CP] Consolidate decode-context-parallel (DCP) helpers into layers/dcp/ (#29365 )

2026-07-03 12:25:39 -07:00

debug_utils

[refactor] Move MLP collective flags onto ForwardFlags (#30802 )

2026-07-10 17:43:39 -07:00

disaggregation

[PD] Improve optimistic prefill (#30951 )

2026-07-12 15:31:53 -07:00

dllm

[dLLM] Make FDFO a framework capability for all dLLM algorithms (#27551 )

2026-07-11 11:05:05 +08:00

dp_attn

Enable breakable prefill CUDA graph for DP attention (#30898 )

2026-07-12 17:10:04 -07:00

dp_engine

[Test] split test/registered/distributed/ into topic folders (#26240 )

2026-05-24 17:02:07 -07:00

[NVIDIA] Support flashinfer a2a with flashinfer_trtllm_routed moe (#22394 )

2026-06-29 16:23:58 -07:00

eplb

LPLB: linear-programming load balancer for MoE expert parallelism (#24515 )

2026-06-16 10:19:42 -07:00

eval

Fix Mistral GSM8K chat eval (#27757 )

2026-07-09 21:08:48 -07:00

function_call

[CI] reduce CPU CI scope with base-c suite (#28623 )

2026-06-24 16:18:06 +08:00

gb300

Update GLM tests to 5.2 and delete redundant tests (#29686 )

2026-06-29 23:04:26 -07:00

hicache

[misc] Use --cuda-graph-max-bs-decode in tests, examples, and docs (#29591 )

2026-06-28 18:38:28 -07:00

input_embedding

[misc] Use --cuda-graph-max-bs-decode in tests, examples, and docs (#29591 )

2026-06-28 18:38:28 -07:00

jit

[Kernel] Migrate scattered quantization kernels to sglang.kernels (RFC #29630 , Phase 2.5, 1/7) (#30784 )

2026-07-13 16:17:01 +08:00

kernels

[Kernel] Migrate scattered quantization kernels to sglang.kernels (RFC #29630 , Phase 2.5, 1/7) (#30784 )

2026-07-13 16:17:01 +08:00

kv_canary

[kv canary] Support UnifiedRadixCache in kv-canary and bracket nested model.forward (#30574 )

2026-07-10 10:58:55 -07:00

layers

Add stochastic rounding for FP16 Mamba SSM cache (#26929 )

2026-06-29 01:47:09 -07:00

lora

[Kernel] Introduce sglang.kernels namespace and migrate scattered triton_ops kernels (RFC #29630 , Phase 2) (#30044 )

2026-07-10 21:41:08 +08:00

mem_cache

Size KV pool after CUDA graph capture (opt-in) (#30157 )

2026-07-07 12:05:01 -07:00

mla

[misc] Use --cuda-graph-max-bs-decode in tests, examples, and docs (#29591 )

2026-06-28 18:38:28 -07:00

mlx/models_e2e

[Apple Silicon] [CI] Move the MLX lane to the check-changes + pr-gate composite (#30121 )

2026-07-10 22:00:15 -07:00

mock_model

[refactor] Read resolved config from server_args fields; retire the flags mirror tier (#30346 )

2026-07-07 21:28:34 -07:00

model_loading

Cleanup: relocate temp_set_env and consolidate multi-device/CUDA helpers in common.py (#30180 )

2026-07-05 18:47:26 -07:00

models

[AMD] Register 2 CPU/ROCm-safe tests for AMD 1-GPU PR CI (#29680 )

2026-07-02 00:25:32 -07:00

models_e2e

[Bugfix] Fix Nemotron ForwardFlags across custom op boundary (#30968 )

2026-07-12 22:10:23 -07:00

moe

[Kernel] Migrate scattered quantization kernels to sglang.kernels (RFC #29630 , Phase 2.5, 1/7) (#30784 )

2026-07-13 16:17:01 +08:00

musa

chore: cleanup garbage code (#29770 )

2026-07-02 16:14:01 +08:00

observability

feat(metrics): add Prometheus metrics for the EPD encoder server (#27564 )

2026-07-07 12:04:51 +08:00

openai_server

fix(anthropic): handle mid-conversation system messages (#26773 )

2026-06-21 04:34:47 +00:00

ops

[Spec] Add DSpark: confidence-scheduled speculative decoding (#30261 )

2026-07-12 17:25:26 -05:00

page_major

feat(mem_cache): page-major (layer-major within a page) KV/state layout (#29533 )

2026-06-29 14:49:54 -07:00

perf

Avoid implicit field-based side channel in Scheduler planning (#29408 )

2026-07-10 08:55:51 +08:00

Add Qwen3 MoE tests for PP compatibility with CP and DP (#29640 )

2026-06-29 18:52:03 +08:00

prefill_only

[CI] Move piecewise CUDA graph (pcg) tests to nightly (#30563 )

2026-07-08 16:07:08 -07:00

profiling

[CI] Rename basic CI stage-a/b/c -> base-a/b/c for symmetry with extra CI (#25420 )

2026-05-15 18:26:55 -07:00

quant

[Kernel] Migrate scattered quantization kernels to sglang.kernels (RFC #29630 , Phase 2.5, 1/7) (#30784 )

2026-07-13 16:17:01 +08:00

radix_cache

[UnifiedTree]: Sync mamba int8 checkpoint (#30626 )

2026-07-10 23:46:38 +08:00

reasoning

[CI] Rename basic CI stage-a/b/c -> base-a/b/c for symmetry with extra CI (#25420 )

2026-05-15 18:26:55 -07:00

Fix DeepEP CI test registration (#30873 )

2026-07-12 05:54:50 -07:00

rotary

[misc] CI hygiene: enforce __main__ entry, drop silent-skipped tests, fix rerun-test protoc (#23305 )

2026-04-20 21:16:24 -07:00

sampling

[CI] Rename basic CI stage-a/b/c -> base-a/b/c for symmetry with extra CI (#25420 )

2026-05-15 18:26:55 -07:00

scheduler

[Spec] Add DSpark: confidence-scheduled speculative decoding (#30261 )

2026-07-12 17:25:26 -05:00

scripted_runtime

[AMD] Register 2 CPU-bound 1-GPU tests (phase_checker, scripted_runtime_core) for AMD PR CI (#30446 )

2026-07-08 14:04:23 -07:00

sessions

[AMD] ci: add extra-a 1-gpu-large tier (fp8kv-triton, streaming-session, spec-standalone) (#28458 )

2026-06-17 23:31:32 -07:00

spec

[Spec] Add DSpark: confidence-scheduled speculative decoding (#30261 )

2026-07-12 17:25:26 -05:00

stress

Add stress test workflow (#13937 )

2025-11-26 13:24:39 -08:00

tokenizer

fix: populate batch req rids and per-request http_worker_ipc for mult… (#29882 )

2026-07-04 12:09:41 -07:00

unit

[mem_cache][7/N] refactor: move MLATokenToKVPoolHost to pool_host.mla (#30616 )

2026-07-13 14:23:38 +08:00

utils

refactor(load-snapshot): build LoadSnapshot directly, drop legacy get_loads IPC (#30525 )

2026-07-09 05:58:10 -07:00

vlm

[misc] Use --cuda-graph-max-bs-decode in tests, examples, and docs (#29591 )

2026-06-28 18:38:28 -07:00

xpu

[XPU] Remove redundant xpu graph backend and make xpu graph opt-in by default (#29911 )

2026-07-03 15:58:56 +08:00

README.md

[CI] Add READMEs for unit test directory structure (#20626 )

2026-03-15 20:27:05 +08:00

README.md

Registered Tests

Tests under this directory are auto-discovered by run_suite.py via CI registration decorators.

Where Should I Put My New Test?

No server / engine launch required

What you're testing	Directory	Requires
Component logic in isolation (cache, scheduler, config, parser, etc.)	`unit/<module>/`	CPU or GPU
CUDA kernel correctness	`kernels/`	GPU

Server / engine launch required (E2E)

What you're testing	Directory	Requires
Model inference correctness	`models/`, `4-gpu-models/`, `8-gpu-models/`	GPU
Feature-specific (OpenAI API, LoRA, speculative, distributed, VLM, etc.)	`openai_server/`, `lora/`, `spec/`, `distributed/`, ...	GPU
Benchmarks (performance, accuracy, stress)	`benchmark/`	GPU
Platform-specific	`amd/`, `ascend/`	Vendor GPU

See unit/README.md for unit test conventions.