sglang

ascend

[Ascend][feature] support L1+ L2 radixcache on ascend (#12214 )

2025-11-12 20:45:24 +08:00

batch_invariant

[Deterministic] Optimize bmm_batch_invariant op (#12522 )

2025-11-04 00:33:31 -08:00

configs

…

cpu

[Fix] Set global args in cpu test (#12105 )

2025-10-24 21:46:17 -07:00

debug_utils

Refactor --debug-tensor-dump-layers to list (#12691 )

2025-11-05 03:30:01 -08:00

entrypoints/http_server

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

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

ep

Super tiny fix typo (#13001 )

2025-11-11 00:47:45 +08:00

function_call

Remove EBNF Composer (#13163 )

2025-11-12 17:55:30 -08:00

hicache

Unify memory management across (overlap, non-overlap) x (page>=1) x (spec, non-spec, spec v2) x (retract, finished) (#12224 )

2025-11-11 02:56:22 +08:00

layers

[MoE] Add Comprehensive MoE Integration Tests (#12090 )

2025-11-07 00:34:46 -08:00

lora

[CI] Fix qwen3-vl lora nightly ci (#12708 )

2025-11-05 11:00:13 -08:00

models

vlm: fix tiny multimodal cache bug (#12984 )

2025-11-10 21:30:19 +08:00

nightly

Re-enable Flashinfer TRTLLM GEN MHA and Add Unit Test (#12885 )

2025-11-10 20:17:43 -08:00

openai_server

[Test] Merge all constrained decoding tests. (#12633 )

2025-11-05 00:43:06 +08:00

quant

[ROCM] Optimized deepseek-r1 model with rmsnorm + fp8 quant fusion (#12689 )

2025-11-11 02:59:10 -08:00

rl

[minor] sync code on python/sglang/test/test_deterministic.py and improve ci tests (#11777 )

2025-10-17 14:25:22 -07:00

rotary_embedding

Support mrope triton kernel and add unit test (#11722 )

2025-10-20 11:51:07 +08:00

xpu

Init attention backend for Intel XPU (#10656 )

2025-10-21 11:41:28 +08:00

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

…

experiment_runner.py

…

kv_cache_scales_llama3_1_8b.json

…

kv_cache_scales_llama3_8b.json

…

kv_cache_scales_qwen2_1_5b.json

…

parse_results.py

misc: parse bench_serving result as markdown table (#9377 )

2025-08-20 16:44:20 -07:00

run_suite.py

[AMD] Add AITER Custom All-Reduce (#13102 )

2025-11-12 21:53:44 -08:00

test_abort.py

feat: return partial generation results when aborting requests in waiting queue (#11673 )

2025-10-29 22:03:00 +08:00

test_async_dynamic_batch_tokenizer.py

[Performance] Dynamic Batch Tokenizer (#9382 )

2025-09-14 01:56:04 +08:00

test_async_mm_data_processor.py

[VLM] Optimize async mm data process mechanism (#12066 )

2025-11-01 01:24:53 +08:00

test_bench_one_batch.py

…

test_bench_serving.py

[ci] fix pp test (#11294 )

2025-10-07 14:20:04 +08:00

test_bnb.py

vlm: enforce pybase64 for image and str encode/decode (#10700 )

2025-10-21 19:05:32 +08:00

test_build_eagle_tree.py

move eagle draft post process to cuda graph (#11434 )

2025-10-14 22:50:53 +08:00

test_chunked_prefill.py

…

test_config_integration.py

chore: Initial support for input config files (#10534 )

2025-09-24 14:45:52 -07:00

test_constrained_decoding.py

[Test] Merge all constrained decoding tests. (#12633 )

2025-11-05 00:43:06 +08:00

test_cpp_radix_cache.py

Tiny simplify evcition metrics collector (#12983 )

2025-11-11 23:23:48 +08:00

test_create_kvindices.py

fix: revert #8593 (#9581 )

2025-08-25 01:29:06 -07:00

test_custom_allreduce.py

[AMD] Add AITER Custom All-Reduce (#13102 )

2025-11-12 21:53:44 -08:00

test_cutedsl_moe.py

[NVIDIA] Fix cutedsl backend of MoE (#12353 )

2025-11-04 18:54:55 -08:00

test_data_parallelism.py

…

test_deepseek_chat_templates.py

Fix DeepSeek chat templates to handle tool call arguments type checking (#11700 ) (#12123 )

2025-10-30 16:39:25 +08:00

test_deepseek_v3_basic.py

Adjust server launch time in ci (#12917 )

2025-11-09 20:41:29 +08:00

test_deepseek_v3_cutedsl_4gpu.py

[NVIDIA] Fix cutedsl backend of MoE (#12353 )

2025-11-04 18:54:55 -08:00

test_deepseek_v3_deterministic.py

[Deterministic] add deepseek v3 deterministic inference CI test (#12412 )

2025-11-01 18:10:32 -07:00

test_deepseek_v3_fp4_4gpu.py

Remove deprecated --enable-beta-spec argument and fix b200 test (#12167 )

2025-10-26 15:54:40 -07:00

test_deepseek_v3_mtp.py

Adjust server launch time in ci (#12917 )

2025-11-09 20:41:29 +08:00

test_deepseek_v32_basic.py

[Deepseek V3.2] Fix accuracy bug in the Indexer (#12583 )

2025-11-11 16:15:26 -08:00

test_deepseek_v32_mtp.py

[Deepseek V3.2] Fix accuracy bug in the Indexer (#12583 )

2025-11-11 16:15:26 -08:00

test_deepseek_v32_nsabackend.py

[Deepseek V3.2] Fix accuracy bug in the Indexer (#12583 )

2025-11-11 16:15:26 -08:00

test_deterministic.py

convert test_deterministic into unit tests (#11095 )

2025-10-06 20:33:11 -07:00

test_disaggregation_basic.py

[PD-Disagg] Check finish after pop tranferred (#12638 )

2025-11-05 11:18:09 +08:00

test_disaggregation_different_tp.py

[ci]use H20 to run disaggregation test (#11543 )

2025-10-16 11:42:42 -07:00

test_disaggregation_dp_attention.py

Refactor / Unify event loop across PD-Disagg, Overlap, DP-Attn cases (#12839 )

2025-11-10 00:42:50 +08:00

test_disaggregation_hybrid_attention.py

[PD] Add PD support for hybrid model (Qwen3-Next, DeepSeek V3.2 Exp) (#10912 )

2025-10-15 18:59:14 -07:00

test_disaggregation_pp.py

[ci]use H20 to run disaggregation test (#11543 )

2025-10-16 11:42:42 -07:00

test_double_sparsity.py

…

test_dp_attention.py

Remove dp balance metadata and minimul token balance. (#11170 )

2025-10-03 01:48:15 +08:00

test_eagle_dp_attention.py

Eagle3 DP attention for Qwen3 MoE (#12002 )

2025-10-29 20:25:17 +08:00

test_eagle_infer_a.py

Tiny move files to utils folder (#11166 )

2025-10-03 22:40:06 +08:00

test_eagle_infer_b.py

…

test_eagle_infer_beta.py

Unify memory management across (overlap, non-overlap) x (page>=1) x (spec, non-spec, spec v2) x (retract, finished) (#12224 )

2025-11-11 02:56:22 +08:00

test_eval_accuracy_large.py

…

test_eval_fp8_accuracy.py

…

test_expert_distribution.py

Move deep gemm related arguments to sglang.srt.environ (#11547 )

2025-10-14 00:34:35 +08:00

test_expert_location_updater.py

…

test_fa3.py

Move deep gemm related arguments to sglang.srt.environ (#11547 )

2025-10-14 00:34:35 +08:00

test_fim_completion.py

Tiny move files to utils folder (#11166 )

2025-10-03 22:40:06 +08:00

test_flash_attention_4.py

[CI] Fix huggingface access for test_flash_attention_4.py (#12846 )

2025-11-07 20:07:06 -08:00

test_flashmla.py

Import flash_mla from sgl-kernel (#12135 )

2025-10-29 23:54:21 -07:00

test_forward_split_prefill.py

Separate allocation logic from scheduler (#11313 )

2025-10-10 17:38:54 -07:00

test_fp4_moe.py

[NVIDIA] Fix unit test of MoE and add it to nightly ci (#12709 )

2025-11-05 14:33:18 -08:00

test_function_call_parser.py

Remove EBNF Composer (#13163 )

2025-11-12 17:55:30 -08:00

test_fused_moe.py

[2/2] Deepseek deterministic: support deepseek v3 deterministic inference on 8 x H200 (#12095 )

2025-10-29 11:49:04 -07:00

test_get_weights_by_name.py

…

test_gguf.py

…

test_gpt_oss_1gpu.py

Fix CI (#9013 )

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

test_gpt_oss_4gpu.py

feat: add gpt oss b200 ci (#9988 )

2025-09-03 17:26:38 -07:00

test_gpt_oss_common.py

Add tests to AMD CI for MI35x (#9662 )

2025-09-10 12:50:05 -07:00

test_gptqmodel_dynamic.py

Depreate global_server_args_dict (#11528 )

2025-10-13 19:34:43 +08:00

test_harmony_parser.py

Move parsers under a single folder (#9912 )

2025-09-02 18:25:04 -07:00

test_health_check.py

…

test_hidden_states.py

…

test_hybrid_attn_backend.py

Move deep gemm related arguments to sglang.srt.environ (#11547 )

2025-10-14 00:34:35 +08:00

test_input_embeddings.py

…

test_io_struct.py

…

test_jinja_template_utils.py

Move parsers under a single folder (#9912 )

2025-09-02 18:25:04 -07:00

test_kv_events.py

…

test_llama31_fp4.py

[CI] Add Llama 3.1 8B FP4 to B200 CI (#12182 )

2025-10-29 14:56:41 -07:00

test_load_weights_from_remote_instance.py

Support loading weights from remote instance (#8215 )

2025-09-12 17:40:22 +08:00

test_local_attn.py

…

test_logprobs.py

feat: limit peak memory usage when computing logprobs (#6318 )

2025-11-03 23:53:20 -08:00

test_mamba_unittest.py

Revert "Enable memory saver for hybrid model" (#12648 )

2025-11-04 16:22:06 -08:00

test_metrics_utils.py

Rename customer label -> custom label (#10899 )

2025-09-25 16:19:53 -07:00

test_metrics.py

…

test_mla_deepseek_v3.py

Use more general heuristics to set the default value of --mem-fraction-static (#10975 )

2025-09-29 10:11:03 -07:00

test_mla_flashinfer.py

fix: revert #8593 (#9581 )

2025-08-25 01:29:06 -07:00

test_mla_fp8.py

…

test_mla_int8_deepseek_v3.py

Fix unit tests (#11503 )

2025-10-12 07:45:57 -07:00

test_mla_tp.py

…

test_mla.py

…

test_modelopt_export.py

Enable native ModelOpt quantization support (3/3) (#10154 )

2025-10-21 21:44:29 -07:00

test_modelopt_fp8kvcache.py

…

test_modelopt_loader.py

Enable native ModelOpt quantization support (3/3) (#10154 )

2025-10-21 21:44:29 -07:00

test_modelopt.py

…

test_models_from_modelscope.py

…

test_moe_eval_accuracy_large.py

…

test_mscclpp.py

…

test_multi_instance_release_memory_occupation.py

Fix mem fraction static for nightly tests (#11076 )

2025-09-29 12:57:41 -07:00

test_multi_tokenizer.py

Support Multi Process Tokenizer Manager(#6555 ) (#8964 )

2025-09-01 01:00:13 -07:00

test_ngram_speculative_decoding.py

Move deep gemm related arguments to sglang.srt.environ (#11547 )

2025-10-14 00:34:35 +08:00

test_no_chunked_prefill.py

…

test_no_overlap_scheduler.py

…

test_original_logprobs.py

Sync code and test CI; rename some env vars (#11686 )

2025-10-15 18:37:03 -07:00

test_page_size.py

…

test_patch_torch.py

Tiny move files to utils folder (#11166 )

2025-10-03 22:40:06 +08:00

test_penalty.py

[Test] Add Functional Tests for Penalty Parameters (#11931 )

2025-11-01 00:10:17 +08:00

test_piecewise_cuda_graph.py

[PieceWise CUDA Graph] Support awq/gptq model in piecewise cudagraph (#12518 )

2025-11-11 11:56:15 +08:00

test_pp_single_node.py

[Fix] Compatibility between DP attention and pipeline parallelism (#10100 )

2025-09-06 01:34:10 -07:00

test_priority_scheduling.py

feat: add priority based scheduling with priority based request acceptance and preemption (#8746 )

2025-09-16 17:10:10 -07:00

test_profile_merger_http_api.py

[FEATURE] Add Profile Trace Merger for Distributed Traces (#11413 )

2025-10-14 09:20:17 +08:00

test_profile_merger.py

[FEATURE] Add Profile Trace Merger for Distributed Traces (#11413 )

2025-10-14 09:20:17 +08:00

test_pytorch_sampling_backend.py

…

test_quantization.py

[9/n] decouple quantization impl from vllm dependency - adjust ci (#12753 )

2025-11-10 14:55:19 -08:00

test_quick_allreduce.py

[bug][rocm]fix qr when variable inp (#11609 )

2025-11-11 01:43:48 -08:00

test_radix_attention.py

[Test] Enhance radix cache test for spec cases (#12394 )

2025-10-30 18:46:28 +08:00

test_radix_cache_unit.py

Fix eagle radix cache (#10846 )

2025-09-30 22:59:20 +08:00

test_reasoning_parser.py

Move parsers under a single folder (#9912 )

2025-09-02 18:25:04 -07:00

test_release_memory_occupation.py

Support releasing CUDA graph memory when paused (#7873 )

2025-10-28 14:40:50 +08:00

test_request_queue_validation.py

Simplify the BatchMultimodalOutput in io_struct.py (#12993 )

2025-11-10 13:54:56 -08:00

test_retract_decode.py

Enhance retract test (page cases, long output cases) (#12781 )

2025-11-11 03:03:26 +08:00

test_rope_rocm.py

Revert "[ROCm] Remove vLLM rope dependency & use AITER impl" (#12028 )

2025-10-23 12:42:59 -07:00

test_sagemaker_server.py

[Test] Handle streaming chunks with null content in case of stream end. (#10862 )

2025-11-12 02:03:50 +08:00

test_schedule_policy.py

feat: add priority based scheduling with priority based request acceptance and preemption (#8746 )

2025-09-16 17:10:10 -07:00

test_score_api.py

[Generative Score API] Multi-Item scoring with custom attention mask. (#10979 )

2025-10-08 18:47:32 -07:00

test_server_args.py

Do not use MagicMock to mock server_args in tests (#12154 )

2025-10-26 16:00:46 +08:00

test_session_control.py

Tiny move files to utils folder (#11166 )

2025-10-03 22:40:06 +08:00

test_skip_tokenizer_init.py

…

test_speculative_registry.py

[Auto Sync] Update scheduler.py, spec_info.py, run_suite.py... (20251027) (#12235 )

2025-10-27 17:21:08 -07:00

test_srt_endpoint.py

[Feature] Add /tokenize and /detokenize OpenAI compatible endpoints (#9545 )

2025-10-08 12:38:48 +08:00

test_srt_engine_with_quant_args.py

…

test_srt_engine.py

Tiny move files to utils folder (#11166 )

2025-10-03 22:40:06 +08:00

test_standalone_speculative_decoding.py

Move deep gemm related arguments to sglang.srt.environ (#11547 )

2025-10-14 00:34:35 +08:00

test_start_profile.py

Fix [test]: Env:SGLANG_TORCH_PROFILER_DIR for pytest. (#10780 )

2025-10-03 22:59:32 +08:00

test_swa_unittest.py

EAGLE cache fix for SWARadixCache (#11231 )

2025-10-07 18:21:37 +08:00

test_tokenizer_batch_encode.py

Tiny move files to utils folder (#11166 )

2025-10-03 22:40:06 +08:00

test_tokenizer_manager.py

Tiny move files to utils folder (#11166 )

2025-10-03 22:40:06 +08:00

test_torch_compile_moe.py

[UT for RL] Add UT to cover release/resume memory case for moe model (#8803 )

2025-09-09 19:25:12 -07:00

test_torch_compile.py

…

test_torch_flex_attention_backend.py

feat: Add FlexAttention Backend for Efficient Sparse Attention (#9947 )

2025-09-18 11:49:17 -07:00

test_torch_native_attention_backend.py

…

test_torch_tp.py

…

test_torchao.py

…

test_tracing.py

[Feature] Sglang Tracing: Fine-Grained Tracking for Request Latency - Part 2 (#10804 )

2025-10-28 01:25:46 -07:00

test_triton_attention_backend.py

…

test_triton_attention_kernels.py

fixes hardcoded "cuda" device references in unit tests to use a dynamic device selection (#12761 )

2025-11-07 11:38:35 +08:00

test_triton_attention_rocm_mla.py

…

test_triton_fused_moe.py

Support true on-policy (#12058 )

2025-10-25 10:23:42 +08:00

test_triton_moe_channel_fp8_kernel.py

Support true on-policy (#12058 )

2025-10-25 10:23:42 +08:00

test_triton_moe_wna16.py

Support true on-policy (#12058 )

2025-10-25 10:23:42 +08:00

test_triton_sliding_window.py

…

test_two_batch_overlap.py

Move deep gemm related arguments to sglang.srt.environ (#11547 )

2025-10-14 00:34:35 +08:00

test_utils_update_weights.py

Fix CI (#9012 )

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

test_vertex_endpoint.py

…

test_vision_chunked_prefill.py

vlm: enforce pybase64 for image and str encode/decode (#10700 )

2025-10-21 19:05:32 +08:00

test_vision_openai_server_a.py

model: support deepseek-ocr (#11891 )

2025-10-24 03:15:17 +08:00

test_vision_openai_server_common.py

model: support deepseek-ocr (#11891 )

2025-10-24 03:15:17 +08:00

test_vlm_accuracy.py

feat: basic support for server-level multimodal cache (#10775 )

2025-11-10 00:27:50 +08:00

test_vlm_input_format.py

Update transformers package version to 4.57.0 (#11222 )

2025-10-05 23:45:14 +00:00

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