Lee Nau 84c8390514 Use dsv3 optimized routing fused_topk_deepseek instead of moe_fused_gate (#15347)

2026-01-19 11:50:16 +08:00

..

ascend

[NPU] NPU quantization refactoring & more quantization formats support (#14504)

2026-01-15 04:25:15 +08:00

configs

…

cpu

…

ep

Re-introduce the unit test of test_mooncake_ep_small (#16019)

2026-01-12 17:01:24 +08:00

models

test: split Qwen3 Next tests and disable PCG tests due to intermittent failures (#16989)

2026-01-13 20:18:42 -08:00

quant

…

xpu

…

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

…

run_suite.py

[Refactor] Add -fp4-gemm-backend to replace SGLANG_FLASHINFER_FP4_GEMM_BACKEND (#16534)

2026-01-18 23:25:46 +08:00

test_bench_one_batch.py

…

test_bench_serving.py

…

test_bench_typebaseddispatcher.py

…

test_bnb.py

…

test_deepseek_v3_basic.py

…

test_deepseek_v3_cutedsl_4gpu.py

…

test_deepseek_v3_fp4_4gpu.py

…

test_deepseek_v3_mtp.py

Use dsv3 optimized routing fused_topk_deepseek instead of moe_fused_gate (#15347)

2026-01-19 11:50:16 +08:00

test_deepseek_v32_basic.py

…

test_deepseek_v32_mtp.py

…

test_disaggregation_aarch64.py

…

test_disaggregation_different_tp.py

…

test_disaggregation_dp_attention.py

…

test_disaggregation_hybrid_attention.py

Fix hybrid attention PD Disaggregation test (#17099)

2026-01-15 23:38:58 +08:00

test_disaggregation_pp.py

…

test_embed_interpolate_unittest.py

…

test_epd_disaggregation.py

…

test_eval_accuracy_large.py

…

test_fla_layernorm_guard.py

…

test_fp8_blockwise_gemm.py

…

test_gguf.py

…

test_gpt_oss_4gpu.py

…

test_gptqmodel_dynamic.py

…

test_int4fp8_moe.py

[AMD][Quantization] Add int4fp8_moe online quantization on ROCm (#7392)

2026-01-14 01:44:40 -08:00

test_mistral_large3_basic.py

…

test_moe_eval_accuracy_large.py

…

test_multi_instance_release_memory_occupation.py

…

test_nvfp4_gemm.py

[Refactor] Add -fp4-gemm-backend to replace SGLANG_FLASHINFER_FP4_GEMM_BACKEND (#16534)

2026-01-18 23:25:46 +08:00

test_pp_single_node.py

…

test_prefill_delayer.py

…

test_profile_v2.py

…

test_quantization.py

…

test_release_memory_occupation.py

…

test_rope_rocm.py

…

test_type_based_dispatcher.py

…