Roberto L. Castro b55d830ec7 [Perf][Kernel] Persistent TopK scheduler: unified CUDAGraph-safe kernel with dynamic per-row dispatch - DeepSeek-V3.2 DSA decode (#37421) ...

Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>

2026-04-08 13:35:57 -04:00

..

attention.yaml

[ci] Sync test areas with test-pipeline.yaml and enable new pipeline generator (#33080)

2026-01-26 12:28:20 -08:00

basic_correctness.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

benchmarks.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

compile.yaml

[mla] Support fused FP8/NVFP4 output quantization in MLA attention (#35792) (#36205)

2026-04-02 21:16:11 -04:00

cuda.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

distributed.yaml

Remove MQ multi-node tests (#38934)

2026-04-03 20:00:08 +00:00

e2e_integration.yaml

[offloader] v2: Hide weight onloading latency via prefetching (#29941)

2026-02-25 17:20:59 -08:00

engine.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

entrypoints.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

expert_parallelism.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

kernels.yaml

[Perf][Kernel] Persistent TopK scheduler: unified CUDAGraph-safe kernel with dynamic per-row dispatch - DeepSeek-V3.2 DSA decode (#37421)

2026-04-08 13:35:57 -04:00

lm_eval.yaml

[Test] E2E Nemotron-3-Super tests (#36803)

2026-03-23 17:49:56 -07:00

lora.yaml

[CI/Build][LoRA] Update Qwen35 LoRA testing (#37816)

2026-03-23 12:55:49 +08:00

misc.yaml

[CI] Add reasoning parser tests to CI (#37025)

2026-04-08 00:57:36 +00:00

model_executor.yaml

[QeRL] Fix online quantized reloading (#38442)

2026-03-29 14:56:41 -06:00

model_runner_v2.yaml

[Model Runner V2] Fuse probabilistic rejection sample kernels (#38496)

2026-04-07 17:37:37 -07:00

models_basic.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

models_distributed.yaml

[Model] Add Phi4ForCausalLMV for microsoft/Phi-4-reasoning-vision-15B (#38306)

2026-04-02 21:14:57 -07:00

models_language.yaml

[release 2.11] Update to torch 2.11 (#34644)

2026-04-07 18:55:48 -07:00

models_multimodal.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

plugins.yaml

[Refactor] Relocate completion and chat completion tests (#37125)

2026-03-17 11:31:23 +08:00

pytorch.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

quantization.yaml

[release 2.11] Update to torch 2.11 (#34644)

2026-04-07 18:55:48 -07:00

ray_compat.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

samplers.yaml

[ROCm][Test] Fix beam search determinism failures from batch-size-dependent FP divergence and removed wrong marker (#34878)

2026-02-19 08:25:26 +00:00

spec_decode.yaml

[ci] Switch some CI jobs to H200 MIG slices (#38956)

2026-04-05 13:26:11 -07:00

weight_loading.yaml

[CI] Temporarily Disable Nightly Failures (#35770)

2026-03-03 01:49:13 +00:00