Files

T

Cheng WanandClaude Fable 5 ceeaec2078 [PD] Support --enable-unified-memory with PD disaggregation (kimi-linear MLA hybrid-Mamba) (#33362 )

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

2026-08-10 16:07:58 -07:00

4-gpu-models

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

8-gpu-models

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

amd

[AMD] Support DeepSeek V4 DSpark on AMD HIP platform (#30964 )

2026-08-08 15:22:14 -07:00

attention

Fix NaN logits from deterministic Triton extend on the unified memory pool (#33517 )

2026-08-10 16:05:04 -07:00

backends

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

bench_fn

[CI] Trim redundant nightly test registrations (#34070 )

2026-08-08 01:42:46 -07:00

chunked_prefill

[VLM] split multimodal scheduling from mm_utils (#32415 )

2026-08-05 20:24:12 +08:00

constrained_decoding

[CI] Fold duplicate-server suites and prune the retract matrix on 1-gpu-5090 (#33745 )

2026-08-05 12:41:51 -07:00

core

[Build] Add srt_empty extra group for device-agnostic install (#31300 )

2026-08-05 13:17:46 -07:00

Install DeepEP from release wheels (#33932 )

2026-08-07 15:38:44 -07:00

cpu

[CI] Bound the CUDA graph capture range in test launches and lift the spec fixture's admission cap (#33776 )

2026-08-05 20:00:30 -07:00

cuda_graph

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

dcp

fix: preserve GQA head mapping in Triton DCP prefill (#34161 )

2026-08-09 18:52:18 -07:00

debug_utils

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

disaggregation

[PD] Support --enable-unified-memory with PD disaggregation (kimi-linear MLA hybrid-Mamba) (#33362 )

2026-08-10 16:07:58 -07:00

dllm

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

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

dp_attn

[Fix] Resolve VLM test image placeholders from the model's own chat template (#33509 )

2026-08-04 02:01:39 -07:00

dp_engine

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

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

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

eplb

[Kernel] Phase 4 batch-3: migrate tangled JIT subsystems + new groups into kernels.ops (RFC #29630 ) (#32045 )

2026-07-22 21:15:03 +08:00

eval

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

function_call

[Kimi K3] Add reasoning, tool-call, and OpenAI serving support (#33025 )

2026-08-01 14:57:23 -07:00

gb300

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

gemm

[Fix] Reject online weight updates while the HPC-Ops router GEMM split cache is active (#31943 )

2026-07-24 19:30:13 +08:00

hicache

[Fix] Release hierarchical cache host pool on graceful shutdown (#31746 )

2026-07-19 22:27:29 -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

kernels

perf(hisparse): fuse the DSv4 value and scale swap-in copy on ROCm (#33484 )

2026-08-10 14:30:15 -07:00

kv_canary

[UnifiedTree]: move /mem_cache/unifed_cache_component dir to /mem_cache/unified_cache (#32484 )

2026-07-28 16:28:21 +08:00

layers

runtime: Add flashinfer rmsnorm + quant fusion support SM90, SM100, SM120- #32994 (#33471 )

2026-08-09 20:15:03 +08:00

lora

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

mem_cache

[CI] Share VLM engines and prune launch matrices on the per-commit H100/H200 suites (#33944 )

2026-08-07 01:22:22 -07:00

mla

[CI] Merge tokenizer worker tests and drop redundant triton attention e2e (#33641 )

2026-08-05 11:55:11 -07:00

mlx/models_e2e

[MLX] Window-bounded SWA KV storage and in-graph sampling (#34166 )

2026-08-09 21:28:51 -07:00

mock_model

[Kernel] Phase 4 batch-3: migrate tangled JIT subsystems + new groups into kernels.ops (RFC #29630 ) (#32045 )

2026-07-22 21:15:03 +08:00

model_loading

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

models_e2e

Fix stale track rows corrupting conv checkpoints under the prefill graph (#34184 )

2026-08-10 08:57:23 +08:00

moe

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

musa

chore: cleanup garbage code (#29770 )

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

npu

[NPU] Improve the execution efficiency and maintainability of pr‑test‑npu (#33724 )

2026-08-08 16:42:44 +08:00

observability

[tracing] sglang tracing v2: support exporting tracing data asynchronously (#30023 )

2026-08-10 15:23:05 +08:00

openai_server

[CI] Pin the rust frontend parity test to eager prefill (#34146 )

2026-08-08 20:07:19 -07:00

ops

config: route parallel config-leaf reads through get_parallel() (#33170 )

2026-08-01 08:58:39 -07:00

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

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

ci: lower VL PP gsm8k threshold to 0.60 (0.65 has zero margin on H100) (#31702 )

2026-07-19 15:41:12 -07:00

prefill_only

[CI] Trim redundant nightly test registrations (#34070 )

2026-08-08 01:42:46 -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

[Feature] Support NVFP4 token embedding in ModelOpt mixed-precision checkpoints (#34222 )

2026-08-09 23:37:54 -07:00

radix_cache

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

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

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

rotary

[rotary] Rebuild the shared RoPE cache entry when its buffers are dead (#33575 )

2026-08-04 17:24:42 -07:00

rust

support rust sglang server (#29799 )

2026-07-31 11:56:31 -07:00

sampling

Deterministic gumbel sampling: clamp u=1 so masked tokens can't be sampled (#33423 )

2026-08-09 21:43:27 +08:00

scheduler

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

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

[CI] Share VLM engines and prune launch matrices on the per-commit H100/H200 suites (#33944 )

2026-08-07 01:22:22 -07:00

spec

[CI] Trim redundant nightly test registrations (#34070 )

2026-08-08 01:42:46 -07:00

stress

…

tokenizer

[CI] Fold duplicate-server suites and prune the retract matrix on 1-gpu-5090 (#33745 )

2026-08-05 12:41:51 -07:00

unit

[PD] Support --enable-unified-memory with PD disaggregation (kimi-linear MLA hybrid-Mamba) (#33362 )

2026-08-10 16:07:58 -07:00

utils

[CI] Trim redundant nightly test registrations (#34070 )

2026-08-08 01:42:46 -07:00

vlm

[CI] Key scheduled CUDA suites by runner_config instead of hand-written jobs (#34186 )

2026-08-09 16:44:53 -07:00

xpu

Fix DeepSeek-OCR batching crash on variable local-crop counts (#33214 )

2026-08-05 18:39:19 -07:00

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.