5 people 2396a61108 [Attention][MLA][DCP] Query replication for MLA decode (DeepSeek-V2/R1 + Kimi-K2.5) (#45964) ...

Signed-off-by: Sungsoo Ha <sungsooh@nvidia.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>

2026-07-20 23:51:27 +00:00

..

test_attention_backends_selection.py

[Model] Remove MiniMaxText01, MiniMaxVL01, MiniMaxForCausalLM (#45993)

2026-06-22 15:20:46 +08:00

test_attention_backends.py

[Attention] FlashAttention 4 SM100 FP8 kv cache support (#42569)

2026-07-20 10:53:27 -07:00

test_attention_splitting.py

[Core] Add MRV2 virtual-batch PCP for MLA (#46570)

2026-07-19 02:53:15 +00:00

test_backend_per_kind.py

[Attention] Allow selecting a different attention backend per KV-cache group (#48012)

2026-07-17 15:19:02 -04:00

test_batch_reordering.py

[Attention] Support distinguishing between short extends and decodes (#37303)

2026-03-20 10:49:36 -07:00

test_chunked_local_attention.py

refactor hard coded device string in test files under tests/v1 and tests/lora (#37566)

2026-04-03 11:21:47 +08:00

test_dspark_noncausal_sparse_mla.py

[1/N] Add dense MHA path for sparse MLA short sequences (#47327)

2026-07-14 00:29:56 +00:00

test_flashinfer_dcp_spec_reorder.py

[ROCm][CI] fix flashinfer import check (#48647)

2026-07-14 20:54:19 +00:00

test_flashinfer_sparse_mla_sm120_api.py

Enable DeepSeek V4 and GLM-5.1 on SM120 (#43477)

2026-06-22 11:54:14 -07:00

test_gdn_metadata_builder.py

[Bugfix] Avoid racy accepted counts in async spec decode (#45100)

2026-06-22 08:53:16 +00:00

test_indexer_dcp_localize.py

[Attention][DSA] support dcp for FLASHINFER_MLA_SPARSE (#46076)

2026-07-01 00:32:20 -04:00

test_indexer_deepseek_v4_slot_mapping.py

[Feat] DeepSeek V4 Rebased (#40860)

2026-04-26 18:31:08 -07:00

test_kv_head_stride_canonicalization.py

[Attention] Make FlexAttention and FlashAttention use num-blocks first layouts (#42095)

2026-05-26 19:55:56 -07:00

test_linear_attention_metadata_builder.py

[Feature] Support MTP speculative decoding for Bailing hybrid models (#44880)

2026-07-06 10:38:50 +08:00

test_mamba_update_block_table.py

[Bugfix] mamba: run single-token extends as decodes (#42430)

2026-05-18 15:26:00 +00:00

test_mla_backends.py

[Attention][MLA][DCP] Query replication for MLA decode (DeepSeek-V2/R1 + Kimi-K2.5) (#45964)

2026-07-20 23:51:27 +00:00

test_mla_prefill_quant_output.py

feat: MLA prefill enable FA4 fp8 output (#43050)

2026-06-16 07:10:59 -07:00

test_mla_prefill_registry.py

[ROCm][V1][MLA] Clone prefill backend state per metadata builder (#46993)

2026-06-30 11:43:54 -07:00

test_mla_prefill_selector.py

[Bugfix] Enable FlashAttention MLA prefill for Mistral Small 4 head dims (#48855)

2026-07-17 18:07:00 +08:00

test_rocm_attention_backends_selection.py

[ROCm][CI] Refresh ROCm base images when docker rocm_base changes (#46904)

2026-07-07 03:10:50 -07:00

test_sparse_mla_backends.py

[Core] Add MRV2 virtual-batch PCP for MLA (#46570)

2026-07-19 02:53:15 +00:00

test_trtllm_attention_integration.py

[2/N][KV-Cache Layout Refactor] Pack K/V into the content dim across attention backends (#44455)

2026-07-11 11:11:16 -04:00

utils.py

[Attention] FlashAttention 4 SM100 FP8 kv cache support (#42569)

2026-07-20 10:53:27 -07:00