roikoren755 067353f67b [Test] Refactor KL divergence and prefix cache branching to kits (#19715)

2026-03-12 16:11:59 +08:00

..

4-gpu-models

[Test] Refactor KL divergence and prefix cache branching to kits (#19715)

2026-03-12 16:11:59 +08:00

8-gpu-models

Fix FA3 swa spec pg_size > 1 (#20369)

2026-03-12 11:42:01 +08:00

amd

[AMD][Mori] Add unittest for mori ep low latency and tbo (#20374)

2026-03-11 22:38:26 -07:00

ascend

[NPU] [Quantization] w4a4 MoE layer support (#18924)

2026-03-11 16:52:35 +03:00

attention

[GDN] Change Attention State Layout from [N, HV, K, V] to [N, HV, V, K] (#20283)

2026-03-12 10:53:12 +08:00

backends

[FlashInfer v0.6.4] [RL] Integrate FlashInfer mxfp8 gemm, MoE, and routed MoE (#19537)

2026-03-10 15:37:57 -07:00

bench_fn

…

constrained_decoding

…

core

[Utils] Add NetworkAddress abstraction for IPv6-safe address handling (#20306)

2026-03-11 00:07:37 -07:00

debug_utils

…

disaggregation

[Performance] Decode Offload improves the long texts performance 100% through dynamic block offload. (#17216)

2026-03-08 17:16:53 +08:00

distributed

[Qwen3.5] Support Qwen3.5 Pipeline Parallelism (#19670)

2026-03-07 23:34:08 +08:00

dllm

…

ep

…

eval

…

function_call

…

hicache

…

kernels

…

layers

Return intermediate Mamba states (#19716)

2026-03-09 16:04:36 +08:00

lora

…

metrics

[Bug] Fix missing TTFT histogram for single-batch requests (#20122)

2026-03-08 15:18:51 -07:00

mla

Relax flaky test thresholds for MLA DeepSeek V3 and AutoRound (#20068)

2026-03-06 17:26:28 -08:00

model_loading

…

models

[Test] Refactor KL divergence and prefix cache branching to kits (#19715)

2026-03-12 16:11:59 +08:00

moe

…

openai_server

feature: support X-Data-Parallel-Rank header to specific dp-rank. (#19832)

2026-03-11 14:53:33 -07:00

ops

[AMD] CI - Add MI35x nightly/PR tests for kv-cache-fp8 and allreduce-fusion (DeepSeek) (#19834)

2026-03-05 07:09:57 -08:00

parser

…

perf

…

piecewise_cuda_graph

[CI] Refactor PCG related CI (#19994)

2026-03-10 14:49:25 +08:00

profiling

…

quant

Relax flaky B200 GSM8K accuracy thresholds (#20304)

2026-03-11 12:35:21 -07:00

radix_cache

[Test] Refactor KL divergence and prefix cache branching to kits (#19715)

2026-03-12 16:11:59 +08:00

rl

[Scheduler] Unify idle checks into is_fully_idle() and fix weight update test (#20296)

2026-03-10 17:50:23 -07:00

rotary

…

sampling

…

scheduler

add mixed chunk unit test and make small refactors (#18776)

2026-03-08 19:56:09 +08:00

sessions

Enable piecewise-cuda-graph when logprob_start_len = -1 (#19453)

2026-03-10 12:50:57 -07:00

spec

Relax flaky B200 GSM8K accuracy thresholds (#20304)

2026-03-11 12:35:21 -07:00

stress

…

tokenizer

…

utils

[Utils] Add NetworkAddress abstraction for IPv6-safe address handling (#20306)

2026-03-11 00:07:37 -07:00

vlm

[VLM] Replace decord with torchcodec for video decoding (#20055)

2026-03-09 19:23:49 +08:00

test_hybrid_dp_ep_tp_mtp.py

…

test_srt_backend.py

…