[Docs] Fix warnings in `mkdocs build` #23649

Zerohertz · 2025-08-26T11:30:15Z

As discussed with @hmellor on Slack, this PR addresses the warnings generated when running mkdocs build. (#22588)

Please let me know if you have any feedback!

Note

Warnings fixed:

WARNING -  griffe: vllm/model_executor/layers/lightning_attn.py:461: No type or annotation for parameter 'q'
WARNING -  griffe: vllm/model_executor/layers/lightning_attn.py:462: No type or annotation for parameter 'k'
WARNING -  griffe: vllm/model_executor/layers/lightning_attn.py:463: No type or annotation for parameter 'v'
WARNING -  griffe: vllm/model_executor/layers/lightning_attn.py:464: No type or annotation for parameter 'ed'
WARNING -  griffe: vllm/model_executor/layers/lightning_attn.py:465: No type or annotation for parameter 'block_size'
WARNING -  griffe: vllm/model_executor/layers/lightning_attn.py:466: No type or annotation for parameter 'kv_history'
WARNING -  griffe: vllm/model_executor/layers/lightning_attn.py:469: No type or annotation for returned value 'output'
WARNING -  griffe: vllm/model_executor/layers/lightning_attn.py:470: No type or annotation for returned value 'kv'

WARNING -  griffe: vllm/model_executor/layers/linear.py:236: No type or annotation for parameter 'bias'
WARNING -  griffe: vllm/model_executor/layers/linear.py:236: Parameter 'bias' does not appear in the function signature
WARNING -  griffe: vllm/model_executor/layers/linear.py:381: No type or annotation for parameter 'output_size'
WARNING -  griffe: vllm/model_executor/layers/linear.py:381: Parameter 'output_size' does not appear in the function signature

WARNING -  griffe: vllm/attention/backends/differential_flash_attn.py:808: No type or annotation for parameter 'query'
WARNING -  griffe: vllm/attention/backends/differential_flash_attn.py:808: Parameter 'query' does not appear in the function signature
WARNING -  griffe: vllm/attention/backends/differential_flash_attn.py:809: No type or annotation for parameter 'key'
WARNING -  griffe: vllm/attention/backends/differential_flash_attn.py:809: Parameter 'key' does not appear in the function signature
WARNING -  griffe: vllm/attention/backends/differential_flash_attn.py:810: No type or annotation for parameter 'value'
WARNING -  griffe: vllm/attention/backends/differential_flash_attn.py:810: Parameter 'value' does not appear in the function signature
WARNING -  griffe: vllm/attention/backends/differential_flash_attn.py:812: Failed to get 'name: description' pair from 'kv_cache = [2, num_blocks, block_size, num_kv_heads, head_size]'

WARNING -  griffe: vllm/engine/async_llm_engine.py:482: No type or annotation for parameter '*args'
WARNING -  griffe: vllm/engine/async_llm_engine.py:483: No type or annotation for parameter '**kwargs'

WARNING -  griffe: vllm/attention/backends/xformers.py:647: No type or annotation for parameter 'output'
WARNING -  griffe: vllm/attention/backends/xformers.py:647: Parameter 'output' does not appear in the function signature
WARNING -  griffe: vllm/attention/backends/xformers.py:477: Failed to get 'name: description' pair from 'kv_cache = [2, num_blocks, block_size * num_kv_heads * head_size]'
WARNING -  griffe: vllm/attention/backends/xformers.py:481: No type or annotation for parameter 'attn_type'
WARNING -  griffe: vllm/attention/backends/xformers.py:481: Parameter 'attn_type' does not appear in the function signature

WARNING -  griffe: vllm/attention/backends/utils.py:563: Failed to get 'exception: description' pair from 'is `None` when required for the calculations.'

WARNING -  griffe: vllm/entrypoints/llm.py:150: No type or annotation for parameter '**kwargs'

WARNING -  griffe: vllm/attention/backends/flash_attn.py:608: Failed to get 'name: description' pair from 'kv_cache = [2, num_blocks, block_size, num_kv_heads, head_size]'
WARNING -  griffe: vllm/attention/backends/flash_attn.py:864: No type or annotation for parameter 'attn_metadata'

WARNING -  griffe: vllm/attention/backends/rocm_flash_attn.py:590: Failed to get 'name: description' pair from 'kv_cache = [2, num_blocks, block_size * num_kv_heads * head_size]'
WARNING -  griffe: vllm/attention/backends/rocm_flash_attn.py:594: No type or annotation for parameter 'attn_type'
WARNING -  griffe: vllm/attention/backends/rocm_flash_attn.py:594: Parameter 'attn_type' does not appear in the function signature

WARNING -  griffe: vllm/sequence.py:150: No type or annotation for parameter 'prompt_token_ids'
WARNING -  griffe: vllm/sequence.py:150: Parameter 'prompt_token_ids' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:151: No type or annotation for parameter 'output_token_ids'
WARNING -  griffe: vllm/sequence.py:151: Parameter 'output_token_ids' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:941: No type or annotation for parameter 'request_id'
WARNING -  griffe: vllm/sequence.py:941: Parameter 'request_id' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:942: No type or annotation for parameter 'is_prompt'
WARNING -  griffe: vllm/sequence.py:942: Parameter 'is_prompt' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:943: No type or annotation for parameter 'seq_data'
WARNING -  griffe: vllm/sequence.py:943: Parameter 'seq_data' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:944: No type or annotation for parameter 'sampling_params'
WARNING -  griffe: vllm/sequence.py:944: Parameter 'sampling_params' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:945: No type or annotation for parameter 'block_tables'
WARNING -  griffe: vllm/sequence.py:945: Parameter 'block_tables' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:947: No type or annotation for parameter 'do_sample'
WARNING -  griffe: vllm/sequence.py:947: Parameter 'do_sample' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:950: No type or annotation for parameter 'token_chunk_size'
WARNING -  griffe: vllm/sequence.py:950: Parameter 'token_chunk_size' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:952: No type or annotation for parameter 'lora_request'
WARNING -  griffe: vllm/sequence.py:952: Parameter 'lora_request' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:953: No type or annotation for parameter 'computed_block_nums'
WARNING -  griffe: vllm/sequence.py:953: Parameter 'computed_block_nums' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:955: No type or annotation for parameter 'state'
WARNING -  griffe: vllm/sequence.py:955: Parameter 'state' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:956: No type or annotation for parameter 'multi_modal_data'
WARNING -  griffe: vllm/sequence.py:956: Parameter 'multi_modal_data' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:957: No type or annotation for parameter 'mm_processor_kwargs'
WARNING -  griffe: vllm/sequence.py:957: Parameter 'mm_processor_kwargs' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:958: No type or annotation for parameter 'encoder_seq_data'
WARNING -  griffe: vllm/sequence.py:958: Parameter 'encoder_seq_data' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:962: No type or annotation for parameter 'cross_block_table'
WARNING -  griffe: vllm/sequence.py:962: Parameter 'cross_block_table' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:1044: No type or annotation for parameter 'parent_seq_id'
WARNING -  griffe: vllm/sequence.py:1044: Parameter 'parent_seq_id' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:1046: No type or annotation for parameter 'output_token'
WARNING -  griffe: vllm/sequence.py:1046: Parameter 'output_token' does not appear in the function signature
WARNING -  griffe: vllm/sequence.py:1047: No type or annotation for parameter 'logprobs'
WARNING -  griffe: vllm/sequence.py:1047: Parameter 'logprobs' does not appear in the function signature

WARNING -  griffe: vllm/entrypoints/openai/tool_parsers/minimax_tool_parser.py:469: No type or annotation for parameter 'args_match'

WARNING -  griffe: vllm/outputs.py:450: Failed to get 'name: description' pair from 'Its length depends on the number of classes.'
WARNING -  griffe: vllm/outputs.py:412: Failed to get 'name: description' pair from 'Its length depends on the hidden dimension of the model.'

WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/entrypoints/llm.py:696: (vllm.LLM.chat) Could not find cross-reference target 'generate'
WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/entrypoints/llm.py:1336: (vllm.LLM.wake_up) Could not find cross-reference target 'sleep'
WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.Sequence'
WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.SequenceGroup'
WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.Sequence'
WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.SequenceGroup'

WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/entrypoints/llm.py:696: (vllm.LLM.chat) Could not find cross-reference target 'generate'
WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/entrypoints/llm.py:1336: (vllm.LLM.wake_up) Could not find cross-reference target 'sleep'
WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.Sequence'
WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.SequenceGroup'
WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.Sequence'
WARNING -  mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.SequenceGroup'
WARNING -  mkdocs_autorefs: api/vllm/engine/llm_engine.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.engine.llm_engine.LLMEngine.add_request) Could not find cross-reference target 'vllm.Sequence'
WARNING -  mkdocs_autorefs: api/vllm/engine/llm_engine.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.engine.llm_engine.LLMEngine.add_request) Could not find cross-reference target 'vllm.SequenceGroup'
WARNING -  mkdocs_autorefs: api/vllm/engine/llm_engine.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.engine.llm_engine.LLMEngine.add_request) Could not find cross-reference target 'vllm.Sequence'
WARNING -  mkdocs_autorefs: api/vllm/engine/llm_engine.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.engine.llm_engine.LLMEngine.add_request) Could not find cross-reference target 'vllm.SequenceGroup'
WARNING -  mkdocs_autorefs: api/vllm/entrypoints/llm.md: from vllm/vllm/entrypoints/llm.py:696: (vllm.entrypoints.llm.LLM.chat) Could not find cross-reference target 'generate'
WARNING -  mkdocs_autorefs: api/vllm/entrypoints/llm.md: from vllm/vllm/entrypoints/llm.py:1336: (vllm.entrypoints.llm.LLM.wake_up) Could not find cross-reference target 'sleep'

WARNING -  griffe: vllm/core/block_manager.py:476: Parameter 'sequence_group' does not appear in the function signature
WARNING -  griffe: vllm/core/block_manager.py:355: Parameter 'sequence_group' does not appear in the function signature
WARNING -  griffe: vllm/core/block_manager.py:408: Parameter 'num_lookahead_slots' does not appear in the function signature
WARNING -  griffe: vllm/core/block_manager.py:423: Parameter 'sequence_group' does not appear in the function signature

WARNING - griffe: vllm/model_executor/layers/lightning_attn.py:461: No type or annotation for parameter 'q' WARNING - griffe: vllm/model_executor/layers/lightning_attn.py:462: No type or annotation for parameter 'k' WARNING - griffe: vllm/model_executor/layers/lightning_attn.py:463: No type or annotation for parameter 'v' WARNING - griffe: vllm/model_executor/layers/lightning_attn.py:464: No type or annotation for parameter 'ed' WARNING - griffe: vllm/model_executor/layers/lightning_attn.py:465: No type or annotation for parameter 'block_size' WARNING - griffe: vllm/model_executor/layers/lightning_attn.py:466: No type or annotation for parameter 'kv_history' WARNING - griffe: vllm/model_executor/layers/lightning_attn.py:469: No type or annotation for returned value 'output' WARNING - griffe: vllm/model_executor/layers/lightning_attn.py:470: No type or annotation for returned value 'kv' Signed-off-by: Zerohertz <[email protected]>

WARNING - griffe: vllm/model_executor/layers/linear.py:236: No type or annotation for parameter 'bias' WARNING - griffe: vllm/model_executor/layers/linear.py:236: Parameter 'bias' does not appear in the function signature WARNING - griffe: vllm/model_executor/layers/linear.py:381: No type or annotation for parameter 'output_size' WARNING - griffe: vllm/model_executor/layers/linear.py:381: Parameter 'output_size' does not appear in the function signature Signed-off-by: Zerohertz <[email protected]>

hmellor

Thanks for this PR! Don't feel you have to do this all in one go. You'll likely run into lots of merge conflicts as main changes if you do.

vllm/model_executor/layers/lightning_attn.py

Zerohertz · 2025-08-26T11:46:13Z

Great, thank you! I will work on this and aim to finalize the PR by tomorrow.

I also have a quick question. To achieve a faster mkdocs build (It taskes almost 5min 😂), I'd like to filter for only the scope I'm interested in. Could you please let me know which part of the mkdocs.yml file I would need to modify?

Sharing that information would be a great help to my work.

Co-authored-by: Harry Mellor <[email protected]> Signed-off-by: Hyogeun Oh (오효근) <[email protected]>

WARNING - griffe: vllm/attention/backends/differential_flash_attn.py:808: No type or annotation for parameter 'query' WARNING - griffe: vllm/attention/backends/differential_flash_attn.py:808: Parameter 'query' does not appear in the function signature WARNING - griffe: vllm/attention/backends/differential_flash_attn.py:809: No type or annotation for parameter 'key' WARNING - griffe: vllm/attention/backends/differential_flash_attn.py:809: Parameter 'key' does not appear in the function signature WARNING - griffe: vllm/attention/backends/differential_flash_attn.py:810: No type or annotation for parameter 'value' WARNING - griffe: vllm/attention/backends/differential_flash_attn.py:810: Parameter 'value' does not appear in the function signature WARNING - griffe: vllm/attention/backends/differential_flash_attn.py:812: Failed to get 'name: description' pair from 'kv_cache = [2, num_blocks, block_size, num_kv_heads, head_size]' Signed-off-by: Zerohertz <[email protected]>

WARNING - griffe: vllm/engine/async_llm_engine.py:482: No type or annotation for parameter '*args' WARNING - griffe: vllm/engine/async_llm_engine.py:483: No type or annotation for parameter '**kwargs' Signed-off-by: Zerohertz <[email protected]>

WARNING - griffe: vllm/attention/backends/xformers.py:647: No type or annotation for parameter 'output' WARNING - griffe: vllm/attention/backends/xformers.py:647: Parameter 'output' does not appear in the function signature WARNING - griffe: vllm/attention/backends/xformers.py:477: Failed to get 'name: description' pair from 'kv_cache = [2, num_blocks, block_size * num_kv_heads * head_size]' WARNING - griffe: vllm/attention/backends/xformers.py:481: No type or annotation for parameter 'attn_type' WARNING - griffe: vllm/attention/backends/xformers.py:481: Parameter 'attn_type' does not appear in the function signature Signed-off-by: Zerohertz <[email protected]>

hmellor · 2025-08-26T12:14:41Z

You might be able to play with

vllm/mkdocs.yaml

Lines 75 to 81 in fdeb3da

    
           - api-autonav: 
        
               modules: ["vllm"] 
        
               api_root_uri: "api" 
        
               exclude: 
        
                 - "re:vllm\\._.*"  # Internal modules 
        
                 - "vllm.third_party" 
        
                 - "vllm.vllm_flash_attn"

to reduce the amount of API reference that is generated by adding things to exclude.

The docs for the API generation package we use is https://github.com/tlambert03/mkdocs-api-autonav. The docs for the package that then renders the generated API docs is https://mkdocstrings.github.io/python/usage/ (this one might be less helpful for reducing the build time but more information can't hurt).

For bonus points, if you see anything that could implement the best practices from https://mkdocstrings.github.io/griffe/guide/users/recommendations/python-code (the package that mkdocs-api-autonav uses to crawl the codebase) that would improve the build time for everyone!

WARNING - griffe: vllm/attention/backends/utils.py:563: Failed to get 'exception: description' pair from 'is `None` when required for the calculations.' Signed-off-by: Zerohertz <[email protected]>

WARNING - griffe: vllm/entrypoints/llm.py:150: No type or annotation for parameter '**kwargs' Signed-off-by: Zerohertz <[email protected]>

Signed-off-by: Zerohertz <[email protected]>

WARNING - griffe: vllm/attention/backends/flash_attn.py:608: Failed to get 'name: description' pair from 'kv_cache = [2, num_blocks, block_size, num_kv_heads, head_size]' WARNING - griffe: vllm/attention/backends/flash_attn.py:864: No type or annotation for parameter 'attn_metadata' Signed-off-by: Zerohertz <[email protected]>

WARNING - griffe: vllm/attention/backends/rocm_flash_attn.py:590: Failed to get 'name: description' pair from 'kv_cache = [2, num_blocks, block_size * num_kv_heads * head_size]' WARNING - griffe: vllm/attention/backends/rocm_flash_attn.py:594: No type or annotation for parameter 'attn_type' WARNING - griffe: vllm/attention/backends/rocm_flash_attn.py:594: Parameter 'attn_type' does not appear in the function signature Signed-off-by: Zerohertz <[email protected]>

Signed-off-by: Zerohertz <[email protected]>

Zerohertz · 2025-08-26T12:47:54Z

vllm/entrypoints/llm.py

        logits_processors: Optional[list[Union[str,
                                               type[LogitsProcessor]]]] = None,
-        **kwargs,
+        **kwargs: Any,


I initially used object for the **kwargs type hint, as I had seen it used in most other codes.
However, this caused issues with mypy, so I changed it to Any.

It looks like you did the right thing. According to https://peps.python.org/pep-0484/#arbitrary-argument-lists-and-default-argument-values you should type hint the type of the expected values of the contents of **kwargs. Since we use **kwargs to pass arbitrary keyword arguments, Any is an apprioriate choice.

WARNING - griffe: vllm/sequence.py:150: No type or annotation for parameter 'prompt_token_ids' WARNING - griffe: vllm/sequence.py:150: Parameter 'prompt_token_ids' does not appear in the function signature WARNING - griffe: vllm/sequence.py:151: No type or annotation for parameter 'output_token_ids' WARNING - griffe: vllm/sequence.py:151: Parameter 'output_token_ids' does not appear in the function signature WARNING - griffe: vllm/sequence.py:941: No type or annotation for parameter 'request_id' WARNING - griffe: vllm/sequence.py:941: Parameter 'request_id' does not appear in the function signature WARNING - griffe: vllm/sequence.py:942: No type or annotation for parameter 'is_prompt' WARNING - griffe: vllm/sequence.py:942: Parameter 'is_prompt' does not appear in the function signature WARNING - griffe: vllm/sequence.py:943: No type or annotation for parameter 'seq_data' WARNING - griffe: vllm/sequence.py:943: Parameter 'seq_data' does not appear in the function signature WARNING - griffe: vllm/sequence.py:944: No type or annotation for parameter 'sampling_params' WARNING - griffe: vllm/sequence.py:944: Parameter 'sampling_params' does not appear in the function signature WARNING - griffe: vllm/sequence.py:945: No type or annotation for parameter 'block_tables' WARNING - griffe: vllm/sequence.py:945: Parameter 'block_tables' does not appear in the function signature WARNING - griffe: vllm/sequence.py:947: No type or annotation for parameter 'do_sample' WARNING - griffe: vllm/sequence.py:947: Parameter 'do_sample' does not appear in the function signature WARNING - griffe: vllm/sequence.py:950: No type or annotation for parameter 'token_chunk_size' WARNING - griffe: vllm/sequence.py:950: Parameter 'token_chunk_size' does not appear in the function signature WARNING - griffe: vllm/sequence.py:952: No type or annotation for parameter 'lora_request' WARNING - griffe: vllm/sequence.py:952: Parameter 'lora_request' does not appear in the function signature WARNING - griffe: vllm/sequence.py:953: No type or annotation for parameter 'computed_block_nums' WARNING - griffe: vllm/sequence.py:953: Parameter 'computed_block_nums' does not appear in the function signature WARNING - griffe: vllm/sequence.py:955: No type or annotation for parameter 'state' WARNING - griffe: vllm/sequence.py:955: Parameter 'state' does not appear in the function signature WARNING - griffe: vllm/sequence.py:956: No type or annotation for parameter 'multi_modal_data' WARNING - griffe: vllm/sequence.py:956: Parameter 'multi_modal_data' does not appear in the function signature WARNING - griffe: vllm/sequence.py:957: No type or annotation for parameter 'mm_processor_kwargs' WARNING - griffe: vllm/sequence.py:957: Parameter 'mm_processor_kwargs' does not appear in the function signature WARNING - griffe: vllm/sequence.py:958: No type or annotation for parameter 'encoder_seq_data' WARNING - griffe: vllm/sequence.py:958: Parameter 'encoder_seq_data' does not appear in the function signature WARNING - griffe: vllm/sequence.py:962: No type or annotation for parameter 'cross_block_table' WARNING - griffe: vllm/sequence.py:962: Parameter 'cross_block_table' does not appear in the function signature WARNING - griffe: vllm/sequence.py:1044: No type or annotation for parameter 'parent_seq_id' WARNING - griffe: vllm/sequence.py:1044: Parameter 'parent_seq_id' does not appear in the function signature WARNING - griffe: vllm/sequence.py:1046: No type or annotation for parameter 'output_token' WARNING - griffe: vllm/sequence.py:1046: Parameter 'output_token' does not appear in the function signature WARNING - griffe: vllm/sequence.py:1047: No type or annotation for parameter 'logprobs' WARNING - griffe: vllm/sequence.py:1047: Parameter 'logprobs' does not appear in the function signature Signed-off-by: Zerohertz <[email protected]>

WARNING - griffe: vllm/entrypoints/openai/tool_parsers/minimax_tool_parser.py:469: No type or annotation for parameter 'args_match' Signed-off-by: Zerohertz <[email protected]>

WARNING - griffe: vllm/outputs.py:450: Failed to get 'name: description' pair from 'Its length depends on the number of classes.' WARNING - griffe: vllm/outputs.py:412: Failed to get 'name: description' pair from 'Its length depends on the hidden dimension of the model.' Signed-off-by: Zerohertz <[email protected]>

WARNING - mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/entrypoints/llm.py:696: (vllm.LLM.chat) Could not find cross-reference target 'generate' WARNING - mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/entrypoints/llm.py:1336: (vllm.LLM.wake_up) Could not find cross-reference target 'sleep' WARNING - mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.Sequence' WARNING - mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.SequenceGroup' WARNING - mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.Sequence' WARNING - mkdocs_autorefs: api/vllm/index.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.LLMEngine.add_request) Could not find cross-reference target 'vllm.SequenceGroup' WARNING - mkdocs_autorefs: api/vllm/engine/llm_engine.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.engine.llm_engine.LLMEngine.add_request) Could not find cross-reference target 'vllm.Sequence' WARNING - mkdocs_autorefs: api/vllm/engine/llm_engine.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.engine.llm_engine.LLMEngine.add_request) Could not find cross-reference target 'vllm.SequenceGroup' WARNING - mkdocs_autorefs: api/vllm/engine/llm_engine.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.engine.llm_engine.LLMEngine.add_request) Could not find cross-reference target 'vllm.Sequence' WARNING - mkdocs_autorefs: api/vllm/engine/llm_engine.md: from vllm/vllm/engine/llm_engine.py:623: (vllm.engine.llm_engine.LLMEngine.add_request) Could not find cross-reference target 'vllm.SequenceGroup' WARNING - mkdocs_autorefs: api/vllm/entrypoints/llm.md: from vllm/vllm/entrypoints/llm.py:696: (vllm.entrypoints.llm.LLM.chat) Could not find cross-reference target 'generate' WARNING - mkdocs_autorefs: api/vllm/entrypoints/llm.md: from vllm/vllm/entrypoints/llm.py:1336: (vllm.entrypoints.llm.LLM.wake_up) Could not find cross-reference target 'sleep' Signed-off-by: Zerohertz <[email protected]>

hmellor · 2025-08-26T13:51:15Z

Some more info to keep in the thread.

As documented in https://mkdocstrings.github.io/python/usage/docstrings/google/#docstring any section that isn't Google style will be transformed into a MkDocs admonition.

WARNING - griffe: vllm/core/block_manager.py:476: Parameter 'sequence_group' does not appear in the function signature WARNING - griffe: vllm/core/block_manager.py:355: Parameter 'sequence_group' does not appear in the function signature WARNING - griffe: vllm/core/block_manager.py:408: Parameter 'num_lookahead_slots' does not appear in the function signature WARNING - griffe: vllm/core/block_manager.py:423: Parameter 'sequence_group' does not appear in the function signature Signed-off-by: Zerohertz <[email protected]>

Zerohertz · 2025-08-26T14:01:16Z

Tip

  - api-autonav:
      modules: ["vllm"]
      api_root_uri: "api"
      exclude:
        - "re:vllm\\._.*"
        - "vllm.third_party"
        - "vllm.vllm_flash_attn"
        - "re:vllm\\.(?!core).*" # This can decrease build time (but can cause auto ref issue, you can ignore it)

Zerohertz · 2025-08-26T14:13:43Z

As you mentioned, I'm also concerned about potential conflicts, so I'll remove the "WIP" and keep the PR open.
I'm done for today and will try to continue working on it tomorrow if I have time.
If this gets merged by then, I'll create a new PR.

Signed-off-by: Harry Mellor <[email protected]>

hmellor · 2025-08-26T15:52:14Z

vllm/sequence.py

-        prompt_token_ids: The token IDs of the prompt.
-        output_token_ids: The token IDs of the output.
-        cumulative_logprob: The cumulative log probability of the output.


I moved these to their respective @property methods because this class seems to have no public attributes

Signed-off-by: Zerohertz <[email protected]> Signed-off-by: Hyogeun Oh (오효근) <[email protected]> Signed-off-by: Harry Mellor <[email protected]> Co-authored-by: Harry Mellor <[email protected]> Signed-off-by: tc-mb <[email protected]>

Signed-off-by: Zerohertz <[email protected]> Signed-off-by: Hyogeun Oh (오효근) <[email protected]> Signed-off-by: Harry Mellor <[email protected]> Co-authored-by: Harry Mellor <[email protected]>

Signed-off-by: Zerohertz <[email protected]> Signed-off-by: Hyogeun Oh (오효근) <[email protected]> Signed-off-by: Harry Mellor <[email protected]> Co-authored-by: Harry Mellor <[email protected]> Signed-off-by: Xiao Yu <[email protected]>

Signed-off-by: Zerohertz <[email protected]> Signed-off-by: Hyogeun Oh (오효근) <[email protected]> Signed-off-by: Harry Mellor <[email protected]> Co-authored-by: Harry Mellor <[email protected]>

Signed-off-by: Zerohertz <[email protected]> Signed-off-by: Hyogeun Oh (오효근) <[email protected]> Signed-off-by: Harry Mellor <[email protected]> Co-authored-by: Harry Mellor <[email protected]> Signed-off-by: Matthew Bonanni <[email protected]>

Signed-off-by: Zerohertz <[email protected]> Signed-off-by: Hyogeun Oh (오효근) <[email protected]> Signed-off-by: Harry Mellor <[email protected]> Co-authored-by: Harry Mellor <[email protected]>

Signed-off-by: Zerohertz <[email protected]> Signed-off-by: Hyogeun Oh (오효근) <[email protected]> Signed-off-by: Harry Mellor <[email protected]> Co-authored-by: Harry Mellor <[email protected]> Signed-off-by: Shiyan Deng <[email protected]>

Signed-off-by: Zerohertz <[email protected]> Signed-off-by: Hyogeun Oh (오효근) <[email protected]> Signed-off-by: Harry Mellor <[email protected]> Co-authored-by: Harry Mellor <[email protected]>

Zerohertz added 2 commits August 26, 2025 20:31

Zerohertz force-pushed the docs/mkdocs-warnings branch from e313771 to 4e192ab Compare August 26, 2025 11:31

hmellor reviewed Aug 26, 2025

View reviewed changes

vllm/model_executor/layers/lightning_attn.py Outdated Show resolved Hide resolved

hmellor self-assigned this Aug 26, 2025

Zerohertz and others added 4 commits August 26, 2025 20:47

docs: type hint in lightning_attn

227ac57

Co-authored-by: Harry Mellor <[email protected]> Signed-off-by: Hyogeun Oh (오효근) <[email protected]>

docs: type hint in async_llm_engine

f1ebf42

WARNING - griffe: vllm/engine/async_llm_engine.py:482: No type or annotation for parameter '*args' WARNING - griffe: vllm/engine/async_llm_engine.py:483: No type or annotation for parameter '**kwargs' Signed-off-by: Zerohertz <[email protected]>

Zerohertz added 4 commits August 26, 2025 21:15

docs: docstring indent in backend utils

94261d8

WARNING - griffe: vllm/attention/backends/utils.py:563: Failed to get 'exception: description' pair from 'is `None` when required for the calculations.' Signed-off-by: Zerohertz <[email protected]>

docs: type hint in entrypoints llm

581a89d

WARNING - griffe: vllm/entrypoints/llm.py:150: No type or annotation for parameter '**kwargs' Signed-off-by: Zerohertz <[email protected]>

fix: import typing.Optional

bd10f66

Signed-off-by: Zerohertz <[email protected]>

mergify bot added the frontend label Aug 26, 2025

Zerohertz added 2 commits August 26, 2025 21:29

fix: kwargs type hints (object -> Any, for CI)

31974c9

Signed-off-by: Zerohertz <[email protected]>

mergify bot added the rocm Related to AMD ROCm label Aug 26, 2025

Zerohertz commented Aug 26, 2025

View reviewed changes

Zerohertz added 4 commits August 26, 2025 21:56

docs: type hint in minimax_tool_parser

7d710fb

WARNING - griffe: vllm/entrypoints/openai/tool_parsers/minimax_tool_parser.py:469: No type or annotation for parameter 'args_match' Signed-off-by: Zerohertz <[email protected]>

mergify bot added the tool-calling label Aug 26, 2025

github-project-automation bot added this to Tool Calling Aug 26, 2025

Zerohertz requested review from comaniac and njhill as code owners August 26, 2025 14:13

Zerohertz changed the title ~~(WIP) [Docs] Fix warnings in mkdocs build~~ [Docs] Fix warnings in mkdocs build Aug 26, 2025

hmellor added 2 commits August 26, 2025 17:47

object -> Any

6d4b1a7

Signed-off-by: Harry Mellor <[email protected]>

Move private attr docstrings to public property methods

87058c2

Signed-off-by: Harry Mellor <[email protected]>

hmellor reviewed Aug 26, 2025

View reviewed changes

hmellor approved these changes Aug 26, 2025

View reviewed changes

Merge branch 'main' into docs/mkdocs-warnings

db59d4d

hmellor enabled auto-merge (squash) August 26, 2025 15:52

github-actions bot added the ready ONLY add when PR is ready to merge/full CI is needed label Aug 26, 2025

hmellor merged commit 730d0ac into vllm-project:main Aug 26, 2025
56 checks passed

github-project-automation bot moved this to Done in Tool Calling Aug 26, 2025

Zerohertz deleted the docs/mkdocs-warnings branch August 27, 2025 00:23

Zerohertz mentioned this pull request Aug 27, 2025

[Docs] Fix warnings in mkdocs build (continued) #23743

Merged

Zerohertz mentioned this pull request Sep 2, 2025

[Docs] Fix warnings in mkdocs build (continued) #24092

Merged

This was referenced Sep 12, 2025

[Docs] Fix warnings in mkdocs build (continued) #24740

Merged

[Docs] Fix warnings in mkdocs build (continued) #24791

Merged

[Docs] Fix warnings in mkdocs build (continued) #25163

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

[Docs] Fix warnings in `mkdocs build` #23649

[Docs] Fix warnings in `mkdocs build` #23649

Uh oh!

Zerohertz commented Aug 26, 2025 •

edited by github-actions bot

Loading

Uh oh!

hmellor left a comment

Uh oh!

Uh oh!

Zerohertz commented Aug 26, 2025 •

edited

Loading

Uh oh!

hmellor commented Aug 26, 2025

Uh oh!

Zerohertz Aug 26, 2025 •

edited

Loading

Uh oh!

hmellor Aug 26, 2025

Uh oh!

hmellor commented Aug 26, 2025

Uh oh!

Zerohertz commented Aug 26, 2025 •

edited

Loading

Uh oh!

Zerohertz commented Aug 26, 2025

Uh oh!

hmellor Aug 26, 2025

Uh oh!

Uh oh!

Uh oh!

Uh oh!

[Docs] Fix warnings in mkdocs build #23649

[Docs] Fix warnings in mkdocs build #23649

Uh oh!

Conversation

Zerohertz commented Aug 26, 2025 • edited by github-actions bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

hmellor left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Zerohertz commented Aug 26, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

hmellor commented Aug 26, 2025

Uh oh!

Zerohertz Aug 26, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

hmellor Aug 26, 2025

Choose a reason for hiding this comment

Uh oh!

hmellor commented Aug 26, 2025

Uh oh!

Zerohertz commented Aug 26, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Zerohertz commented Aug 26, 2025

Uh oh!

hmellor Aug 26, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

[Docs] Fix warnings in `mkdocs build` #23649

[Docs] Fix warnings in `mkdocs build` #23649

Zerohertz commented Aug 26, 2025 •

edited by github-actions bot

Loading

Zerohertz commented Aug 26, 2025 •

edited

Loading

Zerohertz Aug 26, 2025 •

edited

Loading

Zerohertz commented Aug 26, 2025 •

edited

Loading