4 people 3d6853b32c [Bugfix] Seed hybrid mamba state index with mamba_block_size on prefix-cache hits (#55601) ...

Signed-off-by: pondzikk <pondzik@gmail.com>
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: zjy0516 <riverclouds.zhu@qq.com>
Co-authored-by: ptorsten <1652174+ptorsten@users.noreply.github.com>

2026-10-07 17:09:50 +00:00

..

__init__.py

[V1] Adding min tokens/repetition/presence/frequence penalties to V1 sampler (#10681)

2024-12-26 19:02:58 +09:00

test_attn_utils.py

[Bugfix][Attention] Restrict FlashInfer sparse MLA FULL graphs to decode (#59751)

2026-10-06 04:13:23 +00:00

test_cp_utils.py

[PCP][DCP] Support DCP target model with non-DCP Dspark (#56723)

2026-09-26 18:12:35 -04:00

test_dspark_utils.py

[PCP][DCP] Support DCP target model with non-DCP Dspark (#56723)

2026-09-26 18:12:35 -04:00

test_dsv4_packed_zeroer_geometry.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_eagle3_aux_hidden_states_pp.py

[CI] Fix MiniMax-M3 PP aux-state test mock after #58648 (#59508)

2026-09-30 17:34:52 -07:00

test_encoder_runner.py

[MM][V2] Enable encoder-only ViT CUDA graph capture (#56922)

2026-09-17 16:12:17 +08:00

test_gpu_autoregressive_speculator.py

[Bugfix][MRV2][Spec Decode] Honor dynamic K in autoregressive speculators, 1.2~1.3x kernel perf improvement (#57053)

2026-10-05 14:54:11 -04:00

test_gpu_bad_words.py

[Model Runner V2] Support custom logits processors (#56497)

2026-09-19 17:46:49 +00:00

test_gpu_batch_ordering.py

[Perf][PP] Skip sampled-token broadcasts whose requests leave the engine (#58542)

2026-09-30 19:18:55 -07:00

test_gpu_batch_shard.py

[Perf][MRV2] Allow FULL decode graphs for one-token prompt tails (#58400)

2026-09-28 12:48:52 -07:00

test_gpu_block_table.py

[PCP][DCP] Support DCP target model with non-DCP Dspark (#56723)

2026-09-26 18:12:35 -04:00

test_gpu_extract_hidden_states_speculator.py

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

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

test_gpu_gumbel_sample.py

Add pydocstyle to the ruff rules (#52136)

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

test_gpu_input_batch_v2.py

[MRV2] Miscellaneous code cleanup (#57980)

2026-09-23 01:51:52 -07:00

test_gpu_input_batch.py

[Bugfix] V1: clear stale allowed_token_ids mask in InputBatch.condense (#43931)

2026-09-27 03:44:54 +00:00

test_gpu_kpool_tail_slot_mapping.py

[Bugfix][KV Cache][GLM-5.3-Flash] Disable slot mapping kernel for the kpool tail buffer (#57317)

2026-09-17 22:42:54 -04:00

test_gpu_kv_connector.py

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

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

test_gpu_logit_bias.py

[Model Runner V2] Support custom logits processors (#56497)

2026-09-19 17:46:49 +00:00

test_gpu_logits_processors.py

[Model Runner V2] Support custom logits processors (#56497)

2026-09-19 17:46:49 +00:00

test_gpu_model_runner_mm_gather.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_gpu_model_runner_prefill_tail_classification.py

[TEST][XPU][CI] Generalize model runner prefill tail classification tests to all platforms (#60201)

2026-10-07 08:59:20 +08:00

test_gpu_model_runner_v2_cudagraph_profiling.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_gpu_model_runner_v2_eplb.py

[Bugfix][MRV2][Spec Decode] Honor dynamic K in autoregressive speculators, 1.2~1.3x kernel perf improvement (#57053)

2026-10-05 14:54:11 -04:00

test_gpu_model_runner_v2.py

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

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

test_gpu_model_runner.py

[Bugfix][Model Loader] Support reload_weights with runai_streamer load format (#58124)

2026-10-07 14:04:35 +00:00

test_gpu_pcp_manager.py

[Perf][PCP] Shard decode requests across PCP ranks (#52162)

2026-10-01 23:50:13 -04:00

test_gpu_profiler.py

[Metrics] Expose cached prompt tokens by cache tier (#56318)

2026-10-06 14:52:05 -04:00

test_gpu_rejection_sampler_chunking.py

[Feature][Spec Decode] Support sampling mask replay for MRV2 MTP (#59359)

2026-10-01 13:51:39 +08:00

test_gpu_rejection_sampler_i64.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_gpu_sampler_flags.py

[Model Runner V2] Support custom logits processors (#56497)

2026-09-19 17:46:49 +00:00

test_gpu_thinking_budget.py

[Model Runner V2] Support custom logits processors (#56497)

2026-09-19 17:46:49 +00:00

test_gpu_trace_replay.py

[Core][V1] Support trace_decode_token_ids for deterministic decode replay (#46701)

2026-08-20 15:12:43 +08:00

test_gpu_ubatch_slicing.py

[MRV2] Miscellaneous code cleanup (#57980)

2026-09-23 01:51:52 -07:00

test_gpu_warmup_blocks.py

[Bugfix][MRV2][Spec Decode] Honor dynamic K in autoregressive speculators, 1.2~1.3x kernel perf improvement (#57053)

2026-10-05 14:54:11 -04:00

test_gpu_worker_weight_transfer.py

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

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

test_gpu_worker.py

[Bugfix][ROCm] Preserve config during GPU memory profiling (#58014)

2026-10-04 22:28:33 -07:00

test_grammar_invalid_drafts.py

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

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

test_jit_warmup_migration.py

[3/N][Feat][Perf] Add new warmup infrastructure for JITs. Add provider registry and orchestration for JIT warmup (#50174)

2026-08-17 10:11:14 -07:00

test_kv_block_zeroer.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_kv_cache_allocation_scope.py

[5/N] Share HiSparse host cache across TP ranks (reopens #52760) (#56629)

2026-09-12 19:48:25 -04:00

test_late_interaction_runner.py

[Bugfix][Pooling] Handle multimodal cache misses without crashing (#58975)

2026-10-02 10:14:33 -04:00

test_logits_processor_loader.py

[Model Runner V2] Support custom logits processors (#56497)

2026-09-19 17:46:49 +00:00

test_mamba_hybrid_model_state.py

[Bugfix] Seed hybrid mamba state index with mamba_block_size on prefix-cache hits (#55601)

2026-10-07 17:09:50 +00:00

test_mamba_utils.py

[Bugfix] Index Mamba align state copies by request slot in MRV2 (#60210)

2026-10-07 02:32:46 +00:00

test_mixed_warmup_gate.py

[CI] Fix test_mixed_warmup_gate after #57053 (#60116)

2026-10-05 21:17:00 +00:00

test_mrope_prompt_embeds.py

[Security] Fix DoS via prompt_embeds on M-RoPE models (#45252)

2026-06-13 10:17:38 +00:00

test_pp_utils.py

[Perf][PP] Skip sampled-token broadcasts whose requests leave the engine (#58542)

2026-09-30 19:18:55 -07:00

test_prompt_embeds_state.py

[XPU][CI] enable prompt embeds tests on XPU (#58283)

2026-09-25 06:56:59 +08:00

test_sleep_mode_backend.py

[CI] Split (H200 MIG/MI300) Basic Correctness into named jobs (#57054)

2026-09-25 17:05:33 -07:00

test_spec_decode_embed_sharing_pp.py

[Bugfix][MiniMax M3] Share target embeddings with MTP under PP (#58648)

2026-09-29 20:04:10 -07:00

test_utils.py

[HiSparse] Log steady-state max concurrency and expose host-tier utilization gauges (#58949)

2026-10-05 14:20:32 -07:00

test_worker_memory_snapshot.py

Add pydocstyle to the ruff rules (#52136)

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

test_workspace.py

[Core][Kernel] Share persistent workspaces for Marlin and Humming (#57421)

2026-09-20 17:28:33 +08:00

test_xpu_model_runner.py

[XPU] Fix CUDA API shims breaking Torch Dynamo during AOT compile (#43092)

2026-07-06 08:29:20 +08:00