Lianmin Zheng a4825ed588 Fix kernel type annotations for fp8 quant and logging (#16994)

2026-01-13 18:14:32 -08:00

..

ascend

[CI] Migrate nightly tests to test/registered/ (#15582)

2025-12-22 22:16:32 -08:00

attention

Fix Illegal Memory Access when fa3 + spec + topk + page_size > 1 (#15469)

2025-12-24 00:13:57 +08:00

ci

[CI] Enable retry logic for flaky CI tests (#14983)

2025-12-22 22:23:42 -08:00

external_models

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

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

kits

Tiny fix typo minxin -> mixin (#16908)

2026-01-12 00:36:45 +08:00

longbench_v2

Feature/longbench v2 evaluation utils (#10949)

2025-10-07 14:17:31 +08:00

server_fixtures

Tiny print launch command with shlex (#16010)

2025-12-29 11:26:46 +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

fix: nightly fix b200 gpqa (#15745)

2025-12-24 10:29:45 -08:00

bench_one_batch_server_internal.py

Fix kernel type annotations for fp8 quant and logging (#16994)

2026-01-13 18:14:32 -08:00

doc_patch.py

Refactor the docs (#9031)

2025-08-10 19:49:45 -07:00

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

[Auto Sync] Update parallel_state.py, few_shot_gsm8k.py (20250903) (#9986)

2025-09-03 16:55:43 -07:00

get_logits_ut.py

Support DeepSeek V3.2 Exp (#11061)

2025-10-06 00:24:15 -07:00

gpt_oss_common.py

Fix gpt_oss_common import path and migrate core tests (#16426)

2026-01-07 12:58:32 -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

enhance LoRA tests and fix base model LoRA eviction in Scheduler (#16333)

2026-01-10 16:49:00 +08:00

nightly_bench_utils.py

fix: only publish trace from tp 0 (#16411)

2026-01-04 12:17:08 -08:00

nightly_utils.py

[CI] Append test variant name to markdown report header in nightly test (#16166)

2025-12-31 00:09:24 +08:00

performance_test_runner.py

[CI] Append test variant name to markdown report header in nightly test (#16166)

2025-12-31 00:09:24 +08:00

run_combined_tests.py

fix: increasing H200 test timeout (#15600)

2025-12-23 01:00:37 -08:00

run_eval.py

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

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

runners.py

Refactor: separate CI-specific weight validation into dedicated module (#15216)

2025-12-27 20:50:39 -08:00

send_one.py

[Auto Sync] Improve profilers and simplify bench_one_batch_server.py (#13866)

2025-11-25 12:13:31 -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_humaneval.py

Allow benchmarking tool to handle empty response (#12174)

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

simple_eval_longbench_v2.py

[Fix] Improve longbench prompt and other logics (#11474)

2025-10-12 15:04:28 -07:00

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

[AMD] Add silu_and_mul, gelu_and_mul, gelu_tanh_and_mul, and gelu_quick kernels for AMD GPUs (#7135)

2025-07-24 23:44:28 -07:00

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

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

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

test_custom_ops.py

[AMD] Remove vllm's scaled_fp8_quant and moe_sum when SGLANG_USE_AITER=1 (#7484)

2025-07-21 17:33:19 -07:00

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] Improve profilers and simplify bench_one_batch_server.py (#13866)

2025-11-25 12:13:31 -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_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

Replace sglang.srt.layers.quantization.scalar_types with sgl_kernel.scalar_type (#8951)

2025-08-13 19:41:41 -07:00

test_programs.py

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

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

test_utils.py

Tiny support http headers in bench serving (#16606)

2026-01-07 10:15:17 +08:00

vlm_utils.py

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

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