Shangming Cai 384f8ab5ce [PD] Support PD disaggregation with Prefill PP (#8846) ...

Signed-off-by: Shangming Cai <caishangming@linux.alibaba.com>
Signed-off-by: Shangming Cai <csmthu@gmail.com>
Co-authored-by: root <huzhiyuan@xiaohongshu.com>
Co-authored-by: Ying Sheng <sqy1415@gmail.com>
Co-authored-by: Francis <38564764+ssssnow@users.noreply.github.com>
Co-authored-by: zitto <zhjc1124@gmail.com>

2025-08-16 18:31:31 -07:00

..

ascend

Reorganize CI and test files (#9027)

2025-08-10 12:30:06 -07:00

configs

Add test config yamls for Deepseek v3 (#5433)

2025-04-20 17:28:52 -07:00

cpu

[CPU][sgl-kernel] biased_grouped_topk: fix correction_bias dtype to float32 (#8212)

2025-08-04 18:28:31 -07:00

entrypoints/http_server

[RL] Add test for /abort_request (#7626)

2025-08-10 09:14:19 -07:00

ep

Reorganize CI and test files (#9027)

2025-08-10 12:30:06 -07:00

hicache

Reorganize CI and test files (#9027)

2025-08-10 12:30:06 -07:00

lora

Support radix cache for Lora feature (#7216)

2025-08-11 10:14:11 -07:00

models

Reorganize CI and test files (#9027)

2025-08-10 12:30:06 -07:00

openai_server

Fix broken CI TestRequestLengthValidation (#9095)

2025-08-11 22:59:56 -07:00

quant

[CI] add deepseek w4a8 test on h20 ci (#7758)

2025-08-16 01:54:13 -07:00

rl

Support Flatten Tensor Update Weights to speed up MOE Update Weights by 20% (#8079)

2025-08-10 16:08:59 -07:00

double-sparsity-config-Llama-3.1-8B-Instruct.json

Rename double sparsity config file (#2188)

2024-11-25 17:12:08 -08:00

experiment_runner.py

Replace time.time() to time.perf_counter() for benchmarking. (#6178)

2025-05-11 14:32:49 -07:00

kv_cache_scales_llama3_1_8b.json

Support FP8 E4M3 KV Cache (#2786)

2025-01-12 21:17:11 -08:00

kv_cache_scales_llama3_8b.json

support e4m3 kvcache in qwen2 & add kv scaling facotr json (#2894)

2025-01-18 11:43:22 +08:00

kv_cache_scales_qwen2_1_5b.json

support e4m3 kvcache in qwen2 & add kv scaling facotr json (#2894)

2025-01-18 11:43:22 +08:00

parse_results.py

Add test config yamls for Deepseek v3 (#5433)

2025-04-20 17:28:52 -07:00

run_suite.py

[feature] Ascend NPU graph support (#8027)

2025-08-16 17:25:17 -07:00

test_abort.py

Support updating weights at once by stopping all requests (#6698)

2025-07-02 22:26:06 -07:00

test_ascend_graph_tp1_bf16.py

[feature] Ascend NPU graph support (#8027)

2025-08-16 17:25:17 -07:00

test_ascend_graph_tp2_bf16.py

[feature] Ascend NPU graph support (#8027)

2025-08-16 17:25:17 -07:00

test_ascend_w8a8_quantization.py

[Quantization] Supported w8a8 int8 quantized Gemma3 and Qwen-VL models (#8619)

2025-08-12 13:31:18 -07:00

test_bench_one_batch.py

Add V2-lite model test (#7390)

2025-07-03 22:25:50 -07:00

test_bench_serving.py

Reorganize CI and test files (#9027)

2025-08-10 12:30:06 -07:00

test_bnb.py

Implement Siglip Vision model, and support BNB quantization for gemma3-mm (#5339)

2025-05-20 23:53:46 -07:00

test_chunked_prefill.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_cpp_radix_cache.py

[Feature] Radix Tree in C++ (#7369)

2025-08-02 19:50:14 -07:00

test_create_kvindices.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_custom_allreduce.py

[AMD] Add unit-test-sgl-kernel-amd to AMD CI (#7539)

2025-06-29 15:50:09 -07:00

test_data_parallelism.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_disaggregation_different_tp.py

[CI] Fix setup of disaggregation with different tp (#6706)

2025-05-28 11:17:27 -07:00

test_disaggregation_pp.py

[PD] Support PD disaggregation with Prefill PP (#8846)

2025-08-16 18:31:31 -07:00

test_disaggregation.py

[PD] Support decode retract and update decode.py (#7196)

2025-06-14 19:48:05 -07:00

test_double_sparsity.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_dp_attention.py

[CI] Fix CI tests (#9050)

2025-08-10 23:52:05 -07:00

test_eagle_infer_a.py

Improve test cases for eagle infer (#7173)

2025-06-13 22:25:13 -07:00

test_eagle_infer_b.py

[EAGLE] Refactor code for page size > 1 & more simplifications (#7213)

2025-06-16 03:04:29 -07:00

test_ebnf_constrained.py

ci: Fix test_ebnf_generate_all_optional_function_params (#7288)

2025-06-17 16:39:42 -07:00

test_eval_accuracy_large.py

Fix GPU OOM (#6564)

2025-05-24 16:38:39 -07:00

test_eval_fp8_accuracy.py

[CI] Fix ci tests (#5769)

2025-04-27 07:18:10 -07:00

test_expert_distribution.py

Support gathering expert distribution details (#6665)

2025-05-27 15:32:59 -07:00

test_expert_location_updater.py

Move files related to EPLB (#7580)

2025-06-29 15:39:38 -07:00

test_fa3.py

[Eagle] Fix kernel call after updating speculative sampling kernels (#7231)

2025-06-16 07:25:59 -07:00

test_fim_completion.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_flashmla.py

Add V2-lite model test (#7390)

2025-07-03 22:25:50 -07:00

test_forward_split_prefill.py

[Feature] Add a test for Layer-wise Prefill (#8231)

2025-07-21 22:06:15 +08:00

test_full_deepseek_v3.py

Simplify prepare_extend_after_decode (#6987)

2025-06-09 16:39:21 -07:00

test_function_call_parser.py

Support glm4.1v and glm4.5v (#8798)

2025-08-09 00:59:13 -07:00

test_fused_moe.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

test_get_weights_by_name.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_gguf.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_gpt_oss_1gpu.py

Fix CI (#9013)

2025-08-09 16:00:10 -07:00

test_gpt_oss_4gpu.py

Add CI for gpt-oss model on hopper (#8851)

2025-08-09 00:34:23 -07:00

test_gpt_oss_common.py

Fix CI (#9013)

2025-08-09 16:00:10 -07:00

test_gptqmodel_dynamic.py

[3/n] chore: decouple AWQ implementation from vLLM dependency (#8113)

2025-07-18 11:45:22 -07:00

test_health_check.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_hidden_states.py

Open AI API hidden states (#6716)

2025-06-10 14:37:29 -07:00

test_hybrid_attn_backend.py

[feat] Support different attention backends for prefill and decode (#6338)

2025-07-28 11:42:29 +08:00

test_input_embeddings.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_intel_amx_attention_backend.py

Reorganize CI and test files (#9027)

2025-08-10 12:30:06 -07:00

test_io_struct.py

fix: Handles input_embeds in GenerateReqInput when n>1 (#7830)

2025-07-08 14:00:42 -07:00

test_jinja_template_utils.py

Support glm4.1v and glm4.5v (#8798)

2025-08-09 00:59:13 -07:00

test_kv_events.py

feat: add dp-rank to KV events (#6852)

2025-06-04 15:29:34 -07:00

test_local_attn.py

Simplify FA3 tests (#5779)

2025-04-27 01:30:17 -07:00

test_metrics.py

Improve structured outputs: fix race condition, server crash, metrics and style (#6188)

2025-05-11 08:36:16 -07:00

test_mla_deepseek_v3.py

Clean up server_args.py to have a dedicated function for model specific adjustments (#8983)

2025-08-08 19:56:50 -07:00

test_mla_flashinfer.py

Clean up server_args.py to have a dedicated function for model specific adjustments (#8983)

2025-08-08 19:56:50 -07:00

test_mla_fp8.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_mla_int8_deepseek_v3.py

Clean up server_args.py to have a dedicated function for model specific adjustments (#8983)

2025-08-08 19:56:50 -07:00

test_mla_tp.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_mla.py

Improve profiler and integrate profiler in bench_one_batch_server (#6787)

2025-05-31 15:53:55 -07:00

test_modelopt_fp8kvcache.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_modelopt.py

[CI] Fix ci tests (#5769)

2025-04-27 07:18:10 -07:00

test_models_from_modelscope.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_moe_eval_accuracy_large.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_mscclpp.py

support 1 shot allreduce in 1-node and 2-node using mscclpp (#6277)

2025-06-04 22:11:24 -07:00

test_multi_instance_release_memory_occupation.py

Add e2e test for multi instance multi stage memory release/resume occupuation (#7208)

2025-06-26 17:40:38 -07:00

test_nightly_gsm8k_eval_amd.py

Update amd nightly models CI. (#6992)

2025-06-09 10:54:08 -07:00

test_nightly_gsm8k_eval.py

add zai-org/GLM-4.5-Air-FP8 model into nightly CI (#8894)

2025-08-08 01:44:19 -07:00

test_no_chunked_prefill.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_no_overlap_scheduler.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_page_size.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_patch_torch.py

Patch PyTorch's bug that cross-process tensor transfer will lead to wrong device (#4565)

2025-03-27 00:22:33 -07:00

test_penalty.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_pp_single_node.py

[PD] Support PD disaggregation with Prefill PP (#8846)

2025-08-16 18:31:31 -07:00

test_pytorch_sampling_backend.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_quick_allreduce.py

[Feature] Integrate quick allreduce and select the best allreduce implementation (#6619)

2025-07-24 20:48:42 -07:00

test_radix_attention.py

Improve profiler and integrate profiler in bench_one_batch_server (#6787)

2025-05-31 15:53:55 -07:00

test_reasoning_parser.py

Enables force reasoning based on chat template for Qwen3-Thinking (#8369)

2025-08-06 20:02:47 -07:00

test_regex_constrained.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_release_memory_occupation.py

Multi-Stage Awake: Support Resume and Pause KV Cache and Weights separately (#7099)

2025-06-19 00:56:37 -07:00

test_request_queue_validation.py

feat: throttle requests at scheduler based on --max_queued_requests (#7565)

2025-07-28 22:32:33 +08:00

test_retract_decode.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_rope_rocm.py

[AMD] Add test_fused_moe.py and test_rope_rocm.py to AMD CI (#5246)

2025-07-06 01:47:16 -07:00

test_sagemaker_server.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_schedule_policy.py

Fix missing arguments in SchedulePolicy and RadixCache initialization in tests. (#4712)

2025-03-27 22:23:51 -07:00

test_score_api.py

[Generative Score API] Optimization to Remove Decode. (#8840)

2025-08-14 05:12:24 +08:00

test_server_args.py

ci: fix port args (#7792)

2025-07-05 15:06:42 -07:00

test_session_control.py

Move mem_fraction_static adjustment for multimodal models to server_args.py & Fix session control & Other cleanups (#7748)

2025-07-04 16:33:33 -07:00

test_skip_tokenizer_init.py

Support precomputed multimodal features for Qwen-VL and Gemma3 models. (#6136)

2025-05-16 12:26:15 -07:00

test_srt_endpoint.py

[Feature] Add Logit Bias (#6579)

2025-06-10 15:39:25 -07:00

test_srt_engine_with_quant_args.py

Revert "fix some typos" (#6244)

2025-05-12 12:53:26 -07:00

test_srt_engine.py

chore: upgrade flashinfer v0.2.6.post1 jit (#6958)

2025-06-09 09:22:39 -07:00

test_start_profile.py

[feature] Add start step profile argument in /start_profile (#7608)

2025-07-09 18:42:15 -07:00

test_swa_unittest.py

SWA Prefix Cache (#7367)

2025-07-13 12:31:07 -07:00

test_tokenizer_batch_encode.py

[Generative Score API] Optimization to Remove Decode. (#8840)

2025-08-14 05:12:24 +08:00

test_torch_compile_moe.py

Fix throughput threshold for amd ci test (#6414)

2025-05-19 14:17:41 -07:00

test_torch_compile.py

Improve profiler and integrate profiler in bench_one_batch_server (#6787)

2025-05-31 15:53:55 -07:00

test_torch_native_attention_backend.py

[CI] test chunked prefill more (#5798)

2025-04-28 10:57:17 -07:00

test_torch_tp.py

[CI] test chunked prefill more (#5798)

2025-04-28 10:57:17 -07:00

test_torchao.py

Replace time.time() to time.perf_counter() for benchmarking. (#6178)

2025-05-11 14:32:49 -07:00

test_triton_attention_backend.py

[CI] test chunked prefill more (#5798)

2025-04-28 10:57:17 -07:00

test_triton_attention_kernels.py

Add unit test for triton swa kernel (#8853)

2025-08-06 16:10:38 +08:00

test_triton_attention_rocm_mla.py

Add retry for flaky tests in CI (#4755)

2025-03-25 16:53:12 -07:00

test_triton_fused_moe.py

Integrate triton moe kernel (#7689)

2025-07-06 20:05:49 -07:00

test_triton_moe_channel_fp8_kernel.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

test_triton_moe_wna16.py

[6/N] MoE Refactor: Cleanup MoE-related configs (#8849)

2025-08-14 21:14:53 -07:00

test_triton_sliding_window.py

Fix triton sliding window test case (#6981)

2025-06-08 17:20:46 -07:00

test_two_batch_overlap.py

[Feature] improve TBO: two chunk overlap (#8144)

2025-08-05 21:11:01 -07:00

test_utils_update_weights.py

Fix CI (#9012)

2025-08-09 13:33:42 -07:00

test_vertex_endpoint.py

Properly return error response in vertex_generate HTTP endpoint (#5956)

2025-05-01 11:48:58 -07:00

test_vision_chunked_prefill.py

Improve streaming, log_level, memory report, weight loading, and benchmark script (#7632)

2025-06-29 23:16:19 -07:00

test_vision_openai_server_a.py

Fix chat template handling for OpenAI serving (#8635)

2025-07-31 21:49:45 -07:00

test_vision_openai_server_b.py

Support glm4.1v and glm4.5v (#8798)

2025-08-09 00:59:13 -07:00

test_vision_openai_server_common.py

Double vision prefill throughput by defaulting to optimal vision attention backend (#8484)

2025-08-13 02:08:30 -07:00

test_vllm_dependency.py

[CI] Fix flaky threshold (#8370)

2025-07-25 16:41:56 -07:00

test_vlm_accuracy.py

refactor: unify names of the feature field of MultimodalDataItem (#8075)

2025-07-16 17:52:38 -07:00

test_vlm_input_format.py

bring back kimi vl ci (#8537)

2025-07-29 13:14:18 -07:00

test_wave_attention_backend.py

[AMD] Support Wave attention backend with AMD GPU optimizations (#8660)

2025-08-12 13:49:11 -07:00

test_wave_attention_kernels.py

[AMD] Support Wave attention backend with AMD GPU optimizations (#8660)

2025-08-12 13:49:11 -07:00

test_weight_version.py

feat: Add model version tracking with API endpoints and response metadata (#8795)

2025-08-14 12:13:46 -07:00