Juntian LiuandClaude Opus 5.5 388a610dfb [Perf][DFlash] Capture context combine and anchor in the draft CUDA graph (#59511) ...

Signed-off-by: Juntian Liu <Juntianl777@gmail.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

2026-10-14 15:40:52 -04:00

..

__init__.py

Test: added acceptance length tests (#32030)

2026-01-20 18:55:15 +00:00

test_acceptance_estimator.py

[Model Runner V2] Acceptance estimation for adaptive verification (#52228)

2026-09-14 15:58:48 -07:00

test_acceptance_length.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

test_adaptive_verification.py

Disable mypy arg-type and assignment checks in tests (#59428)

2026-09-30 15:18:50 +00:00

test_dflash2.py

Disable mypy arg-type and assignment checks in tests (#59428)

2026-09-30 15:18:50 +00:00

test_dflash_causality.py

[Model] Add Cohere2MoE Eagle3 auxiliary hidden states (#49819)

2026-09-17 20:09:36 +00:00

test_dflash_lookahead.py

Disable mypy arg-type and assignment checks in tests (#59428)

2026-09-30 15:18:50 +00:00

test_dflash_prepare_inputs.py

[Bugfix] Fix DSpark graph-padding handling in the DeepSeek-V4.1 MoE fused top-k (#59166)

2026-10-14 07:49:52 +08:00

test_dflash_profile.py

[Bugfix][Spec Decode] Cap DFlash/DSpark profiling query batch (#56448)

2026-09-22 12:36:32 +08:00

test_draft_attention_backend_override.py

[Fast Start] Cache the MTP draft model in a separate daemon group (#57312)

2026-09-22 03:13:38 +00:00

test_draft_moe_backend_override.py

[Fast Start] Cache the MTP draft model in a separate daemon group (#57312)

2026-09-22 03:13:38 +00:00

test_dspark_topk.py

[Spec Decode] Add NVFP4 DSpark gathered top-k projection (#55713)

2026-09-09 17:33:46 +00:00

test_dsv4_mtp_local_argmax.py

Disable mypy arg-type and assignment checks in tests (#59428)

2026-09-30 15:18:50 +00:00

test_dynamic_sd_cug.py

[MyPy][3/N] Fix MyPy errors in test groups (part 3) (#55939)

2026-09-30 11:06:07 +00:00

test_dynamic_sd.py

Disable dynamic speculative decoding when DP is enabled (#45963)

2026-07-07 12:52:41 +00:00

test_eagle_draft_attn_metadata.py

[GLM 5.3 Perf] Enable fused multi-step decode, 13.3% E2E throughput improvement for concurrency 1 (#57443)

2026-10-02 20:05:27 -04:00

test_eagle_step_kernel.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212)

2026-09-16 10:57:24 -07:00

test_eagle.py

[KV Cache] GLM-5.3-Flash: use the generic packed KV layout; CircularBufferSpec tail with speculative ring slots (#57169)

2026-10-09 15:48:37 -04:00

test_extract_hidden_states.py

Disable mypy arg-type and assignment checks in tests (#59428)

2026-09-30 15:18:50 +00:00

test_lilicorr.py

[Perf][DFlash] Capture context combine and anchor in the draft CUDA graph (#59511)

2026-10-14 15:40:52 -04:00

test_llm_base_proposer_sampling.py

Disable mypy arg-type and assignment checks in tests (#59428)

2026-09-30 15:18:50 +00:00

test_llm_base_proposer.py

[KV Cache] GLM-5.3-Flash: use the generic packed KV layout; CircularBufferSpec tail with speculative ring slots (#57169)

2026-10-09 15:48:37 -04:00

test_max_len.py

[ModelRunner V2] Speculative Decoding NGram GPU Implementations (#40704)

2026-10-06 13:32:59 -07:00

test_mtp_structured_output.py

[Bugfix][Structured Output] Feed the token that implicitly ends reasoning to the grammar (#59627)

2026-10-08 19:38:33 +00:00

test_mtp.py

[Model] Use upstream GLM-5.3 and Qwen4-Exp configs and processor (#57387)

2026-10-02 02:25:18 +00:00

test_ngram_gpu.py

[MRV2] Miscellaneous code structure updates (#60754)

2026-10-09 17:58:09 -07:00

test_ngram.py

[V1][Spec Decode] Add Dynamic SD (#32374)

2026-06-14 00:14:27 -07:00

test_rejection_sampler_utils.py

[Bugfix] Don't let a structured-output request sample from an unmasked row (#54442)

2026-10-01 02:11:44 +00:00

test_request_acceptance.py

[Frontend][Core][Spec Decode] Per-request acceptance stats in OpenAI API responses (#48915)

2026-08-20 15:42:49 -07:00

test_speculators_correctness.py

[Cleanup] Remove online quantization support in fp8.py in favor of online shorthands (#53585)

2026-09-24 12:48:14 -07:00

test_speculators_eagle3.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

test_synthetic_rejection_sampler_utils.py

[Feat] Unified Synthetic Acceptance Rate for V1 and V2 (#40662)

2026-04-24 00:48:08 +00:00

test_vocab_mapping.py

[Perf] Remove CPU-GPU sync in heterogeneous vocabulary speculative decoding (#57396)

2026-09-22 18:34:19 -04:00