Skip to content

Conversation

@ilyasher
Copy link
Contributor

@ilyasher ilyasher commented Nov 21, 2025

Add MoE collector for VLLM and collect data for H100

Perf results

I only collected h100 data for now. It is poor compared to TRTLLM. While TRTLLM is about ~30% of calculated SOL, the collected VLLM data is <10% of SOL. I suspect this is because the TRTLLM data collector ran autotune for MoE kernels, but vllm did not. vllm does have a autotuner for MoE, but it takes hours per test case, so it is not feasible to run it for our data collection.

VLLM MoE sanity check
moe_updated

TRTLLM for reference
Screenshot 2025-11-20 at 7 54 50 PM

@copy-pr-bot
Copy link

copy-pr-bot bot commented Nov 21, 2025

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@ilyasher ilyasher changed the title Add MoE support for VLLM feat: Add MoE support for VLLM Nov 21, 2025
@ilyasher ilyasher marked this pull request as draft November 21, 2025 04:12
@github-actions github-actions bot added the feat label Nov 21, 2025
@ilyasher ilyasher force-pushed the dev-isherstyuk-vllm-moe branch from 5921c78 to b7db778 Compare November 21, 2025 19:59
@github-actions
Copy link

filepath function $$\textcolor{#23d18b}{\tt{passed}}$$ $$\textcolor{#f14c4c}{\tt{failed}}$$ SUBTOTAL
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_argument\_parsing.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIArgumentParsing.test\_supported\_systems\_parse\_successfully}}$$ $$\textcolor{#23d18b}{\tt{2}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{2}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_argument\_parsing.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIArgumentParsing.test\_hf\_id\_as\_alternative\_to\_model}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_argument\_parsing.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIArgumentParsing.test\_model\_and\_hf\_id\_are\_mutually\_exclusive}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_argument\_parsing.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIArgumentParsing.test\_default\_values\_are\_set}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_integration.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIIntegration.test\_cli\_main\_success\_flow}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_argument\_parsing.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIArgumentParsing.test\_debug\_mode\_flag}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_integration.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIIntegration.test\_cli\_main\_success\_flow\_exp\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_argument\_parsing.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIArgumentParsing.test\_save\_directory\_argument}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_integration.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIIntegration.test\_cli\_main\_build\_dispatch}}$$ $$\textcolor{#23d18b}{\tt{2}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{2}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_argument\_parsing.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIArgumentParsing.test\_optional\_parameters}}$$ $$\textcolor{#23d18b}{\tt{5}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{5}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_integration.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIIntegration.test\_cli\_main\_unsupported\_mode\_raises}}$$ $$\textcolor{#23d18b}{\tt{2}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{2}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_integration.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIIntegration.test\_cli\_main\_runtime\_failure}}$$ $$\textcolor{#23d18b}{\tt{2}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{2}}$$
$$\textcolor{#23d18b}{\tt{tests/cli/workflow/test\_argument\_parsing.py}}$$ $$\textcolor{#23d18b}{\tt{TestCLIArgumentParsing.test\_decode\_system\_defaults\_to\_system}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestContextAttention.test\_query\_context\_attention\_sol\_full\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestContextMLA.test\_query\_context\_mla\_different\_tp\_sizes}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestContextAttention.test\_query\_context\_attention\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_base\_queries.py}}$$ $$\textcolor{#23d18b}{\tt{test\_query\_nccl\_sol\_mode\_alltoall\_and\_reduce\_scatter}}$$ $$\textcolor{#23d18b}{\tt{2}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{2}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_base\_queries.py}}$$ $$\textcolor{#23d18b}{\tt{test\_query\_p2p\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_base\_queries.py}}$$ $$\textcolor{#23d18b}{\tt{test\_system\_spec\_was\_loaded\_correctly}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_get\_database\_with\_yaml\_and\_data\_path}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_get\_all\_databases}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_custom\_allreduce\_data\_nonexistent}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_custom\_allreduce\_data\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_nccl\_data\_nonexistent}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_nccl\_data\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_gemm\_data\_nonexistent}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_gemm\_data\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_moe\_data\_nonexistent}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_moe\_data\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_context\_attention\_data\_nonexistent}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_context\_attention\_data\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_generation\_attention\_data\_nonexistent}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_generation\_attention\_data\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_context\_mla\_data\_nonexistent}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_context\_mla\_data\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_generation\_mla\_data\_nonexistent}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_generation\_mla\_data\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_mla\_bmm\_data\_nonexistent}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_data\_loaders.py}}$$ $$\textcolor{#23d18b}{\tt{test\_load\_mla\_bmm\_data\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_database\_helpers.py}}$$ $$\textcolor{#23d18b}{\tt{test\_get\_supported\_databases\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_database\_helpers.py}}$$ $$\textcolor{#23d18b}{\tt{test\_get\_supported\_databases\_empty\_dir}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_database\_helpers.py}}$$ $$\textcolor{#23d18b}{\tt{test\_get\_supported\_databases\_edge\_cases}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_database\_helpers.py}}$$ $$\textcolor{#23d18b}{\tt{test\_get\_latest\_database\_version\_prefers\_stable\_over\_rc}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_database\_helpers.py}}$$ $$\textcolor{#23d18b}{\tt{test\_get\_latest\_database\_version\_rc\_only}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_database\_helpers.py}}$$ $$\textcolor{#23d18b}{\tt{test\_get\_latest\_database\_version\_nonexistent\_returns\_none}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_database\_helpers.py}}$$ $$\textcolor{#23d18b}{\tt{test\_get\_latest\_database\_version\_unparseable\_versions}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_database\_helpers.py}}$$ $$\textcolor{#23d18b}{\tt{test\_get\_latest\_database\_version\_major\_version\_rc\_is\_newer}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestContextAttention.test\_query\_context\_attention\_non\_sol\_mode\_mha}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestGenerationMLA.test\_query\_generation\_mla\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestNcclEdgeCases.test\_query\_nccl\_non\_sol\_single\_gpu}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestContextAttention.test\_query\_context\_attention\_non\_sol\_mode\_xqa}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestGenerationMLA.test\_query\_generation\_mla\_non\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestNcclEdgeCases.test\_query\_nccl\_non\_sol\_interpolation}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestContextAttention.test\_query\_context\_attention\_assertion\_error}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestGenerationMLA.test\_query\_generation\_mla\_sol\_full\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestNcclEdgeCases.test\_query\_nccl\_non\_sol\_large\_gpu\_count}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestGenerationAttention.test\_query\_generation\_attention\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{test\_default\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_base\_queries.py}}$$ $$\textcolor{#23d18b}{\tt{test\_query\_gemm\_exact\_match}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_base\_queries.py}}$$ $$\textcolor{#23d18b}{\tt{test\_query\_custom\_allreduce\_sol\_mode\_calculation}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_base\_queries.py}}$$ $$\textcolor{#23d18b}{\tt{test\_query\_custom\_allreduce\_sol\_full\_returns\_full\_tuple}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_base\_queries.py}}$$ $$\textcolor{#23d18b}{\tt{test\_query\_custom\_allreduce\_non\_sol\_mode\_uses\_custom\_latency}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_base\_queries.py}}$$ $$\textcolor{#23d18b}{\tt{test\_query\_nccl\_sol\_mode\_all\_gather}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestNcclEdgeCases.test\_query\_nccl\_edge\_message\_sizes}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestGenerationAttention.test\_query\_generation\_attention\_sol\_full\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestGemmInterpolation.test\_query\_gemm\_extrapolation}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestDatabaseCache.test\_get\_database\_caching}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestDatabaseCache.test\_get\_database\_no\_data\_path}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestAllreduceEdgeCases.test\_query\_custom\_allreduce\_single\_gpu}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestGenerationAttention.test\_query\_generation\_attention\_non\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestSupportedQuantModes.test\_supported\_quant\_modes\_structure}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestAllreduceEdgeCases.test\_query\_custom\_allreduce\_large\_tp\_scaling}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestGenerationAttention.test\_query\_generation\_attention\_non\_sol\_mode\_mha}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestSupportedQuantModes.test\_supported\_quant\_modes\_values}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestAllreduceEdgeCases.test\_query\_custom\_allreduce\_extrapolation}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestInitializationEdgeCases.test\_extrapolation\_during\_init}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestGenerationAttention.test\_query\_generation\_attention\_edge\_cases}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_edge\_cases.py}}$$ $$\textcolor{#23d18b}{\tt{TestGemmInterpolation.test\_query\_gemm\_interpolation}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestInterpolationMethods.test\_nearest\_1d\_point\_helper\_inner}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestContextMLA.test\_query\_context\_mla\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestInterpolationMethods.test\_interp\_1d}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestInterpolationMethods.test\_nearest\_1d\_point\_helper\_outer}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_attention.py}}$$ $$\textcolor{#23d18b}{\tt{TestContextMLA.test\_query\_context\_mla\_non\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestInterpolationMethods.test\_bilinear\_interpolation}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestInterpolationMethods.test\_nearest\_1d\_point\_helper\_errors}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestExtrapolateDataGrid.test\_extrapolate\_data\_grid\_boundary\_extension}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestInterpolationMethods.test\_interp\_3d\_linear}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestInterpolationMethods.test\_validate}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestCorrectData.test\_correct\_gemm\_data}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestInterpolationMethods.test\_interp\_2d\_1d}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMoE.test\_query\_moe\_different\_workload\_distributions}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestCorrectData.test\_correct\_generation\_attention\_data}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestInterpolationMethods.test\_interp\_3d}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMoE.test\_query\_moe\_edge\_cases}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestExtrapolateDataGrid.test\_extrapolate\_data\_grid\_basic}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestUpdateSupportMatrix.test\_support\_matrix\_creation}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMLABMM.test\_query\_mla\_bmm\_sol\_mode\_pre}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMoE.test\_query\_moe\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestExtrapolateDataGrid.test\_extrapolate\_data\_grid\_with\_sqrt}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMLABMM.test\_query\_mla\_bmm\_sol\_mode\_post}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMoE.test\_query\_moe\_sol\_full\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_interpolation.py}}$$ $$\textcolor{#23d18b}{\tt{TestExtrapolateDataGrid.test\_extrapolate\_data\_grid\_edge\_cases}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMLABMM.test\_query\_mla\_bmm\_sol\_full\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMoE.test\_query\_moe\_non\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMLABMM.test\_query\_mla\_bmm\_non\_sol\_mode\_post}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMLABMM.test\_query\_mla\_bmm\_non\_sol\_mode\_pre}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestSupportedModels.test\_specific\_models\_are\_supported}}$$ $$\textcolor{#23d18b}{\tt{4}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{4}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestSupportedModels.test\_model\_configs\_have\_correct\_structure}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestSupportedModels.test\_model\_moe\_detection}}$$ $$\textcolor{#23d18b}{\tt{4}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{4}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestHFModelSupport.test\_supported\_hf\_models\_exists}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestHFModelSupport.test\_hf\_models\_map\_to\_valid\_model\_configs}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestHFModelSupport.test\_hf\_id\_resolves\_to\_correct\_model\_family}}$$ $$\textcolor{#23d18b}{\tt{4}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{4}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestHFModelSupport.test\_hf\_id\_moe\_detection}}$$ $$\textcolor{#23d18b}{\tt{4}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{4}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestBackendConfiguration.test\_backend\_enum\_exists}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestBackendConfiguration.test\_default\_backend\_is\_trtllm}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestQuantizationModes.test\_gemm\_quant\_modes\_exist}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestQuantizationModes.test\_attention\_quant\_modes\_exist}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestQuantizationModes.test\_moe\_quant\_modes\_exist}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_agg\_default}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_disagg\_default}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_profile\_application}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_total\_gpus\_limits\_agg\_workers}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_agg\_yaml\_patch\_overrides}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_yaml\_file\_profiles\_and\_patch}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_disagg\_profile\_patch\_expands\_replica}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_disagg\_total\_gpus\_caps\_replica}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_disagg\_total\_gpus\_with\_patch}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_disagg\_wideep\_expands\_lists}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_agg\_total\_gpus\_negative\_rejected}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskconfig\_disagg\_total\_gpus\_small\_rejected}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_taskrunner\_runs\_agg\_and\_disagg}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/task/test\_task.py}}$$ $$\textcolor{#23d18b}{\tt{test\_sglang\_moe\_configs}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/test\_common.py}}$$ $$\textcolor{#23d18b}{\tt{TestSupportedSystems.test\_supported\_systems\_exists}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/test\_common.py}}$$ $$\textcolor{#23d18b}{\tt{TestSupportedSystems.test\_supported\_systems\_matches\_yaml\_files\_and\_folders}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/test\_utils.py}}$$ $$\textcolor{#23d18b}{\tt{TestParseHFConfig.test\_parse\_llama\_config}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/test\_utils.py}}$$ $$\textcolor{#23d18b}{\tt{TestParseHFConfig.test\_parse\_moe\_config}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/test\_utils.py}}$$ $$\textcolor{#23d18b}{\tt{TestParseHFConfig.test\_parse\_deepseek\_config}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/test\_utils.py}}$$ $$\textcolor{#23d18b}{\tt{TestParseHFConfig.test\_parse\_config\_with\_head\_dim}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/test\_utils.py}}$$ $$\textcolor{#23d18b}{\tt{TestGetModelConfigFromHFID.test\_successful\_download}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/test\_utils.py}}$$ $$\textcolor{#23d18b}{\tt{TestSafeMkdir.test\_safe\_mkdir\_exists}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestP2P.test\_query\_p2p\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMLABMM.test\_query\_mla\_bmm\_different\_configs}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestP2P.test\_query\_p2p\_sol\_full\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMemoryOperations.test\_query\_mem\_op\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestP2P.test\_query\_p2p\_non\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestP2P.test\_query\_p2p\_edge\_cases}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/models/test\_model\_config.py}}$$ $$\textcolor{#23d18b}{\tt{TestSupportedModels.test\_supported\_models\_list\_exists}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMemoryOperations.test\_query\_mem\_op\_sol\_full\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMemoryOperations.test\_query\_mem\_op\_non\_sol\_mode}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#23d18b}{\tt{tests/sdk/database/test\_moe\_mla.py}}$$ $$\textcolor{#23d18b}{\tt{TestMemoryOperations.test\_query\_mem\_op\_edge\_cases}}$$ $$\textcolor{#23d18b}{\tt{1}}$$ $$\textcolor{#666666}{\tt{0}}$$ $$\textcolor{#23d18b}{\tt{1}}$$
$$\textcolor{#f14c4c}{\tt{tests/cli/e2e\_validation/test\_e2e\_sweep.py}}$$ $$\textcolor{#f14c4c}{\tt{TestE2ESweep.test\_e2e\_configuration\_sweep}}$$ $$\textcolor{#23d18b}{\tt{4}}$$ $$\textcolor{#f14c4c}{\tt{1}}$$ $$\textcolor{#f14c4c}{\tt{5}}$$
$$\textcolor{#f14c4c}{\tt{TOTAL}}$$ $$\textcolor{#23d18b}{\tt{175}}$$ $$\textcolor{#f14c4c}{\tt{1}}$$ $$\textcolor{#f14c4c}{\tt{176}}$$

@ilyasher ilyasher marked this pull request as ready for review November 22, 2025 00:12
@github-actions github-actions bot requested a review from xueh-nv November 22, 2025 00:12
@tianhaox
Copy link
Contributor

  1. I think we can move forward with this sanity check. Need to do operation-level alignment with real inference in future task.
  2. if moe for vllm cannot support etp > 1, we need to write this logic to enumerate_parallel_config() in sdk/pareto_analysis.py to filter out these cases for vllm.

davilu-nvidia and others added 3 commits November 25, 2025 13:56
Signed-off-by: Ilya Sherstyuk <[email protected]>
Signed-off-by: Ilya Sherstyuk <[email protected]>
@ilyasher ilyasher force-pushed the dev-isherstyuk-vllm-moe branch from ff4e819 to 2f5b658 Compare November 25, 2025 21:57
@ilyasher ilyasher changed the base branch from main to dev-isherstyuk-factor-out-test-cases November 25, 2025 21:59
@ilyasher ilyasher requested a review from a team as a code owner November 25, 2025 21:59
@ilyasher ilyasher changed the base branch from dev-isherstyuk-factor-out-test-cases to dev-isherstyuk-refactor-moe-helper-funcs November 25, 2025 22:00
Signed-off-by: Ilya Sherstyuk <[email protected]>
@ilyasher ilyasher force-pushed the dev-isherstyuk-vllm-moe branch from 67ea728 to c209445 Compare November 25, 2025 22:03
@ilyasher
Copy link
Contributor Author

I found my mistake when ep>1, so I updated the collector and re-collected the h100 moe perf. I also updated enumerate_parallel_config() as suggested by @tianhaox

@ilyasher ilyasher changed the title feat: Add MoE support for VLLM (do not merge) feat: Add MoE support for VLLM Nov 25, 2025
@ilyasher ilyasher changed the title (do not merge) feat: Add MoE support for VLLM feat: Add MoE support for VLLM Nov 26, 2025
@ilyasher ilyasher merged commit b17ba13 into dev-isherstyuk-refactor-moe-helper-funcs Nov 26, 2025
6 of 7 checks passed
ilyasher added a commit that referenced this pull request Nov 26, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants