Files

T

Chao ShiandZhangheng bf9773e1da HiCache: Add @rank_consensus to various functions (#37425 )

Co-authored-by: Zhangheng <hzh0425@apache.org>

2026-09-14 10:44:21 +08:00

accuracy/models

[Test] Consolidate test cleanup and CI taxonomy (net -11.4K lines) (#37436 )

2026-09-07 15:13:59 +08:00

amd

[AMD] Fix Dspark accept length and reduce host bubble on DSV4 (#39116 )

2026-09-12 15:21:43 -07:00

attention

[AMD][CI] Skip failing Wave test and relax multi-LoRA output check (#38585 )

2026-09-11 12:25:20 -07:00

backends

[NVIDIA] Support flashinfer Mega Moe (#31470 )

2026-09-10 00:22:47 -07:00

bench_fn

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

chunked_prefill

Fix multimodal embedding cache retaining full batches through views (#39120 )

2026-09-12 21:37:14 -07:00

constrained_decoding

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

core

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

cpu

[Config] msgspec.Struct for the config tier (#38753 )

2026-09-09 19:41:19 -07:00

cuda_graph

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

dcp

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

debug_utils

[XPU] Adapt device agnostic API usage (#32093 )

2026-09-11 10:03:12 +08:00

disaggregation

[unified-memory] PD disaggregation for every unified pool shape (#37506 )

2026-09-13 19:16:43 -07:00

dllm

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

dp_attn

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

dp_engine

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

e2e

[unified-memory] Enable prefill cuda-graph capture (#37418 )

2026-09-13 19:17:19 -07:00

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

eplb

config: delete the redundant full stamp in initialize_model_parallel (#39202 )

2026-09-12 17:27:21 -07:00

expert_pack

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

function_call

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

gb300

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

gemm

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

hicache

HiCache: Add @rank_consensus to various functions (#37425 )

2026-09-14 10:44:21 +08:00

input_embedding

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

jit

[Diffusion] Fuse FLUX.2 gated residual normalization on Blackwell (#37112 )

2026-09-01 08:38:33 +08:00

kernel

[Qwen3.8] Enable NVIDIA NVFP4 on DGX Spark with file-backed PLE and PDL router fix (#39126 )

2026-09-13 16:23:41 +08:00

kernels

[Diffusion] Optimize Qwen-Image-Edit attention on Hopper (#38584 )

2026-09-13 09:15:01 +08:00

kv_canary

Fix KV-canary workspace accounting after graph capture (#38596 )

2026-09-10 16:10:12 -07:00

layers

Fix Qwen3.5 GDN multi-item scoring (#33922 )

2026-09-09 22:30:39 -07:00

lora

[LoRA] Support MoE in full and breakable prefill CUDA graphs (#38578 )

2026-09-11 14:40:02 -07:00

mem_cache

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

mla

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

mlx/models_e2e

[Test] Consolidate test cleanup and CI taxonomy (net -11.4K lines) (#37436 )

2026-09-07 15:13:59 +08:00

mock_model

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

model_loading

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

moe

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

musa

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

npu

[NPU] Set DEEPEP_HYBRID_DEPLOYMENT for new DeepEP tests; switch glm5_2 to w8a8; tune nightly timeouts (#38775 )

2026-09-10 23:32:54 +08:00

observability

[AMD] Restore AMD CI registrations dropped by #37436 (#38581 )

2026-09-09 23:59:10 -07:00

openai_server

[AMD] Restore AMD CI registrations dropped by #37436 (#38581 )

2026-09-09 23:59:10 -07:00

ops

[CI][RFC] Replace black-jupyter with ruff-format (#37210 )

2026-09-02 19:46:08 -07:00

page_major

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

perf

[Test] Consolidate test cleanup and CI taxonomy (net -11.4K lines) (#37436 )

2026-09-07 15:13:59 +08:00

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

prefill_only

[Config] msgspec.Struct for the config tier (#38753 )

2026-09-09 19:41:19 -07:00

profiling

[Test] Consolidate test cleanup and CI taxonomy (net -11.4K lines) (#37436 )

2026-09-07 15:13:59 +08:00

quant

[CI] Drop the GPTQ dynamic-config test for the deleted non-Marlin kernel (#38881 )

2026-09-10 20:55:06 +08:00

radix_cache

HiCache: Add @rank_consensus to various functions (#37425 )

2026-09-14 10:44:21 +08:00

reasoning

[AMD] Restore AMD CI registrations dropped by #37436 (#38581 )

2026-09-09 23:59:10 -07:00

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

rotary

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

rust

[Rust TreeCore] Harden runtime and CI parity (#37303 )

2026-09-09 10:40:22 +08:00

sampling

[Sampling] Support sampling masks with overlap scheduling (#36631 )

2026-09-10 14:52:44 -07:00

scheduler

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

scripted_runtime

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

sessions

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

spec

[NPU] Support DFlash speculative decoding for MiMo-V2.5-Pro (mxfp4) (#37565 )

2026-09-12 11:59:23 +08:00

stress/models

[Test] Consolidate test cleanup and CI taxonomy (net -11.4K lines) (#37436 )

2026-09-07 15:13:59 +08:00

tokenizer

chore: update CI test est_time values (#38238 )

2026-09-06 17:49:41 -07:00

unit

[NPU] Support DSV4 host memory cache management (#37382 )

2026-09-14 10:38:50 +08:00

utils

[Intel][XPU] Add NUMA node binding support for Intel XPU (#31113 )

2026-09-07 10:39:21 +08:00

vlm

[Config] Retire get_global_server_args, and clear the deprecated flags that have a replacement (#38375 )

2026-09-08 16:42:12 -07:00

xpu

Speculative Decoding support for intel_xpu attention backend on XPU target (#30548 )

2026-09-11 10:25:25 +08:00

README.md

Add pr tests (#31952 )

2026-08-01 15:03:21 +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/`, `npu/`	Vendor GPU

See unit/README.md for unit test conventions.