sglang

ascend

[NPU] update nightly tests (#17952 )

2026-02-03 00:13:30 +08:00

attention

fix: fix MockModelRunner in attention tests (#18240 )

2026-02-04 13:18:02 -08:00

ci

feat: add cuda core dump CI warpper (#18909 )

2026-02-17 14:49:26 -08:00

external_models

Fix external_models import path and migrate model loading tests (#16458 )

2026-01-08 23:43:49 -08:00

kits

Enhance bench_multiturn.py with OpenAI API support and richer metrics (#19724 )

2026-03-03 13:48:04 -08:00

longbench_v2

…

server_fixtures

[CI] feat: add early exit to wait_for_server when process dies (#18602 )

2026-02-13 16:46:09 -08:00

speculative

Support FlashAttention3 page_size > 1 and topk > 1 case with paged attn and spec decode (#7725 )

2025-11-26 11:44:41 +08:00

__init__.py

…

accuracy_test_runner.py

Enable SGLANG_ENABLE_SPEC_V2 for nightly speculative decoding tests (#18719 )

2026-02-14 23:00:33 +08:00

bench_one_batch_server_internal.py

[AMD][with CI Fix] support two batch overlapping for mori ep (#19216 )

2026-02-25 02:14:08 -08:00

doc_patch.py

…

few_shot_gsm8k_engine.py

[Fix] Fix uvloop get_event_loop() is not suitable for 0.22.x (#13612 )

2025-11-25 01:20:00 +08:00

few_shot_gsm8k.py

[Misc] Normalize --host parameter to use plain hostname without scheme (#19309 )

2026-02-25 00:37:24 -08:00

get_logits_ut.py

…

gpt_oss_common.py

[AMD] rocm 7.2 image release, PR test, Nightly Test (#17799 )

2026-02-11 21:29:25 -08:00

kl_test_utils.py

[Qwen3-next] support mamba radix cache for overlap scheduler (#14792 )

2025-12-14 18:54:16 -08:00

long_prompt.txt

…

lora_utils.py

[TestFix] change LoRA tests to use NVIDIA adapter instead of Nutanix (#19642 )

2026-03-02 12:55:41 -08:00

nightly_bench_utils.py

Upload nightly test metrics to GH artifacts (#17696 )

2026-01-25 14:35:14 -08:00

nightly_utils.py

Fix eval tests not capturing server launch failures (#18886 )

2026-02-18 07:59:03 +08:00

performance_test_runner.py

Enable SGLANG_ENABLE_SPEC_V2 for nightly speculative decoding tests (#18719 )

2026-02-14 23:00:33 +08:00

run_combined_tests.py

Improve error output in tnightly tets (#18053 )

2026-01-31 19:26:19 -08:00

run_eval.py

[NVIDIA] Add --top-k argument to run_eval.py (#18025 )

2026-02-02 22:17:53 -08:00

runners.py

[args] Add Expert Parallelism Argument To SRT Runner (#18492 )

2026-03-03 14:16:35 -08:00

send_one.py

Refactor sampler: Use a better hash function for deterministic sampling and clear dispatch for probs/logprobs/logits sampling paths (#18915 )

2026-02-17 15:41:23 -08:00

simple_eval_aime25.py

[Feature] Add AIME25 dataset support for SGLang simple_eval (#14990 )

2025-12-15 21:59:40 -08:00

simple_eval_common.py

Add top-p to run_eval.py (#16844 )

2026-01-10 17:10:37 +08:00

simple_eval_gpqa.py

[Lint] Add python/sglang to ruff F401 checks and remove unused imports in files (#11685 )

2025-10-17 16:49:46 -07:00

simple_eval_gsm8k.py

[eval] GSM8k support for run_eval (#17041 )

2026-01-16 11:10:17 +08:00

simple_eval_humaneval.py

Allow benchmarking tool to handle empty response (#12174 )

2025-10-29 20:26:23 -07:00

simple_eval_longbench_v2.py

…

simple_eval_math.py

Allow benchmarking tool to handle empty response (#12174 )

2025-10-29 20:26:23 -07:00

simple_eval_mgsm.py

…

simple_eval_mmlu.py

Allow benchmarking tool to handle empty response (#12174 )

2025-10-29 20:26:23 -07:00

simple_eval_mmmu_vlm.py

fix nightly vlm ci : restore original eval for requests without regex (#14875 )

2025-12-10 23:13:25 -08:00

test_activation.py

…

test_block_fp8_deep_gemm_blackwell.py

[1/2] Refactor DeepGeem requant for FP8 Linear on Blackwell (#13601 )

2025-11-23 16:07:56 -08:00

test_block_fp8.py

[WIP]enable mxfp8 on nvidia sm120 (#19112 )

2026-03-01 19:06:43 -08:00

test_custom_ops.py

…

test_cutlass_moe.py

[sgl-kernel][5/N]Support Expert Specialization Grouped GEMM (#12666 )

2025-11-11 21:23:25 -08:00

test_cutlass_w4a8_moe.py

[Lint] Add python/sglang to ruff F401 checks and remove unused imports in files (#11685 )

2025-10-17 16:49:46 -07:00

test_cutlass_w16a16_moe.py

Add support for bf16 x bf16 cutlass fused MoE (#10275 )

2025-11-24 18:49:39 -08:00

test_deepep_utils.py

…

test_deterministic_utils.py

[Auto Sync] Update test_deterministic.py, test_deterministi... (20251024) (#12083 )

2025-10-24 22:28:01 -07:00

test_deterministic.py

[Auto Sync] Update test_deterministic.py (20260214) (#18839 )

2026-02-14 17:19:30 -08:00

test_dump_metric.py

Add unified metrics collection framework (v1) (#16064 )

2026-01-03 16:30:38 -08:00

test_dynamic_grad_mode.py

…

test_flashinfer_dispatcher.py

[NVIDIA] Add flashinfer all-to-all MOE dispatcher (#14668 )

2026-01-24 22:59:55 +08:00

test_http_server_auth.py

feat: add --admin-api-key for finer-grained endpoint auth (#15908 )

2026-01-13 20:21:55 -08:00

test_kvfp4_quant_dequant.py

feat: Add FP4 (E2M1) KV Cache Support with Quantization Utilities for MLA (#10078 )

2025-11-01 22:24:58 -07:00

test_layernorm.py

Enable mixed type LayerNorm kernel for NSA indexer (#12044 )

2025-11-03 16:50:41 -08:00

test_marlin_utils.py

…

test_programs.py

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

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

test_utils.py

feat(benchmark script): add similar to vllm --ready-check-timeout-sec parameter (#15466 )

2026-03-03 13:44:38 -08:00

tool_call_test_runner.py

Fix parallel tool call test for speculative decoding variants (#19370 )

2026-02-26 18:31:20 -08:00

vlm_utils.py

ci: migrate VLM tests to test/registered/vlm/ (#16415 )

2026-01-05 21:25:29 -08:00