Skip to content

[Misc] Remove unnecessary detokenization in multimodal processing #12868

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Feb 7, 2025
Merged

[Misc] Remove unnecessary detokenization in multimodal processing #12868

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
547781a
Remove unnecessary detokenization in multimodal processing
DarkLight1337 Feb 7, 2025
971c340
Fix
DarkLight1337 Feb 7, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions tests/entrypoints/openai/test_audio.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -83,7 +83,7 @@ async def test_single_chat_session_audio(client: openai.AsyncOpenAI,
choice = chat_completion.choices[0]
assert choice.finish_reason == "length"
assert chat_completion.usage == openai.types.CompletionUsage(
completion_tokens=10, prompt_tokens=202, total_tokens=212)
completion_tokens=10, prompt_tokens=201, total_tokens=211)

message = choice.message
message = chat_completion.choices[0].message
Expand Down Expand Up @@ -140,7 +140,7 @@ async def test_single_chat_session_audio_base64encoded(
choice = chat_completion.choices[0]
assert choice.finish_reason == "length"
assert chat_completion.usage == openai.types.CompletionUsage(
completion_tokens=10, prompt_tokens=202, total_tokens=212)
completion_tokens=10, prompt_tokens=201, total_tokens=211)

message = choice.message
message = chat_completion.choices[0].message
Expand Down Expand Up @@ -196,7 +196,7 @@ async def test_single_chat_session_input_audio(
choice = chat_completion.choices[0]
assert choice.finish_reason == "length"
assert chat_completion.usage == openai.types.CompletionUsage(
completion_tokens=10, prompt_tokens=202, total_tokens=212)
completion_tokens=10, prompt_tokens=201, total_tokens=211)

message = choice.message
message = chat_completion.choices[0].message
Expand Down
4 changes: 2 additions & 2 deletions tests/entrypoints/openai/test_vision.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -92,7 +92,7 @@ async def test_single_chat_session_image(client: openai.AsyncOpenAI,
choice = chat_completion.choices[0]
assert choice.finish_reason == "length"
assert chat_completion.usage == openai.types.CompletionUsage(
completion_tokens=10, prompt_tokens=775, total_tokens=785)
completion_tokens=10, prompt_tokens=774, total_tokens=784)

message = choice.message
message = chat_completion.choices[0].message
Expand Down Expand Up @@ -185,7 +185,7 @@ async def test_single_chat_session_image_base64encoded(
choice = chat_completion.choices[0]
assert choice.finish_reason == "length"
assert chat_completion.usage == openai.types.CompletionUsage(
completion_tokens=10, prompt_tokens=775, total_tokens=785)
completion_tokens=10, prompt_tokens=774, total_tokens=784)

message = choice.message
message = chat_completion.choices[0].message
Expand Down
4 changes: 2 additions & 2 deletions tests/entrypoints/openai/test_vision_embedding.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -93,5 +93,5 @@ async def test_image_embedding(server: RemoteOpenAIServer, model_name: str,
assert len(embeddings.data) == 1
assert len(embeddings.data[0].embedding) == 3072
assert embeddings.usage.completion_tokens == 0
assert embeddings.usage.prompt_tokens == 764
assert embeddings.usage.total_tokens == 764
assert embeddings.usage.prompt_tokens == 763
assert embeddings.usage.total_tokens == 763
3 changes: 0 additions & 3 deletions vllm/inputs/preprocess.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -260,9 +260,6 @@ def _process_multimodal(
mm_processor = self.mm_registry.create_processor(
self.model_config, tokenizer)

if isinstance(prompt, list):
prompt = tokenizer.decode(prompt)

if mm_processor_kwargs is None:
mm_processor_kwargs = {}

Expand Down