3 people d608dfabfd [Bugfix][MRV2] Reserve spec-decode lookahead blocks in V2 warmup (#51438) ...

Signed-off-by: Rahul Chalamala <22563365+rchalamala@users.noreply.github.com>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Rahul Chalamala <22563365+rchalamala@users.noreply.github.com>
Co-authored-by: Janelle Cai <janelle.cai@modal.com>

2026-08-08 23:58:43 +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

[Spec Decode] Support mixed KV page sizes for DFlash (#45181)

2026-06-21 22:45:14 +08:00

test_cp_utils.py

DCP supports hybrid attention (#40996)

2026-07-09 21:34:45 -07:00

test_dsv4_packed_zeroer_geometry.py

[Test] Add packed DeepSeek-V4 KV zeroer geometry regression (#51288)

2026-08-07 12:11:09 -07:00

test_encoder_runner.py

[Bugfix][EPD][Model Runner V2] Skip gather mm embeddings for encoder only instance (#51222)

2026-08-06 20:40:47 -07:00

test_gpu_autoregressive_speculator.py

[Bugfix][Model Runner V2] Restore multimodal draft capability detection (#50417)

2026-08-03 14:18:14 -07:00

test_gpu_batch_ordering.py

[Bugfix][Model Runner V2] Order uniform decodes first so spec decodes aren't misclassified as prefills (#47381)

2026-07-09 14:26:27 +01:00

test_gpu_block_table.py

[BugFix] Stop dummy runs from writing mamba state through stale block-table rows (#49757)

2026-07-29 01:09:41 +00:00

test_gpu_gumbel_sample.py

[Model Runner V2] Cache draft logits in model's LM head dtype (#50910)

2026-08-05 19:12:46 -07:00

test_gpu_input_batch_v2.py

[XPU][CI] Use platform device in InputBatch V2 test (#49939)

2026-07-27 15:28:02 +08:00

test_gpu_input_batch.py

DCP supports hybrid attention (#40996)

2026-07-09 21:34:45 -07:00

test_gpu_model_runner_mm_gather.py

Encoder cache extension hooks (#48218)

2026-07-24 02:52:00 -07:00

test_gpu_model_runner_v2_eplb.py

feat: extended EPLB support for Mistral Large 3 and additional MoE backends (#48355)

2026-08-07 05:32:52 -07:00

test_gpu_model_runner.py

[CI] Detect and fail evals on when NaNs appear in logits (#50323)

2026-08-04 23:02:58 +00:00

test_gpu_profiler.py

[Hardware][GPU] Profiler config additional to increase it scope and annotation details (#37524)

2026-07-17 13:38:59 -07:00

test_gpu_rejection_sampler_chunking.py

[MRV2][Spec Decode] Avoid rejection sampler OOM by chunking (#48630)

2026-07-23 18:13:13 +08:00

test_gpu_rejection_sampler_i64.py

[Bugfix][Model Runner V2][Spec Decode] Fix int32 offset overflow in block verification kernels (#47383)

2026-07-02 07:32:38 -07:00

test_gpu_thinking_budget.py

[Feat] Support thinking_token_budget in Model Runner V2 (#46727)

2026-08-07 03:36:18 +00:00

test_gpu_warmup_blocks.py

[Bugfix][MRV2] Reserve spec-decode lookahead blocks in V2 warmup (#51438)

2026-08-08 23:58:43 +00:00

test_gpu_worker_weight_transfer.py

[Bugfix] Fix level-2 sleep/wake/reload with enable_lora=True (#39935)

2026-08-06 09:13:10 +08:00

test_gpu_worker.py

[Misc] Move PyNvVideoCodec stuff out of gpu worker (#49322)

2026-07-22 10:11:56 +08:00

test_kv_block_zeroer.py

[Bugfix] Fix packed KV block zeroing stride (#50276)

2026-08-06 13:49:06 -07:00

test_late_interaction_runner.py

[Core] Move MRV1 late_interaction_runner.py out of MRV2 subtree (#48014)

2026-07-08 18:30:11 +00:00

test_mamba_hybrid_model_state.py

[ModelRunnerV2] Fix scalar Mamba state update with int32 mappings (#50327)

2026-08-03 22:05:12 +00:00

test_mamba_utils.py

[Hybrid] Stage the postprocess inputs with a single loop over the request list (#48120)

2026-08-03 20:24:22 +00:00

test_mixed_warmup_gate.py

[BugFix] Gate MRV2 mixed sparse-MLA warmup on max_num_seqs > 1 (#47050)

2026-06-30 16:31:27 +01:00

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_sleep_mode_backend.py

[Core] Make sleep-mode backend capability flags communicator-agnostic (#47243)

2026-07-01 07:17:44 +00:00

test_utils.py

[V0 Deprecation] Refactor kv cache from list to element (#37487)

2026-03-23 20:10:11 -07:00

test_worker_memory_snapshot.py

[Hardware] Replace torch.cuda.device_count/current_device/set_device API (#36145)

2026-03-12 07:57:47 -07:00

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