Skip to content
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.

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

[Frontend] Generate valid tool call IDs when using tokenizer-mode=mistral #12332

Merged
merged 19 commits into from
Feb 12, 2025
Merged
Show file tree
Hide file tree
Changes from 1 commit
Commits
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: 6 additions & 0 deletions vllm/entrypoints/openai/serving_chat.py
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,7 @@
from vllm.outputs import CompletionOutput, RequestOutput
from vllm.sampling_params import BeamSearchParams, SamplingParams
from vllm.sequence import Logprob
from vllm.utils import generate_valid_mistral_tool_id
from vllm.transformers_utils.tokenizer import AnyTokenizer, MistralTokenizer
from vllm.transformers_utils.tokenizers import maybe_serialize_tool_calls

Expand Down Expand Up @@ -710,6 +711,11 @@
"completion.")
message = ChatMessage(role=role, content=output.text)

if isinstance(tokenizer, MistralTokenizer):
for tool_call in message.tool_calls:
tool_call.id = generate_valid_mistral_tool_id()
logger.warning(f"Assigned new tool_id: {tool_call.id} for tool: {tool_call}")

Check failure on line 718 in vllm/entrypoints/openai/serving_chat.py

View workflow job for this annotation

GitHub Actions / pre-commit

Ruff (G004)

vllm/entrypoints/openai/serving_chat.py:718:25: G004 Logging statement uses f-string

Check failure on line 718 in vllm/entrypoints/openai/serving_chat.py

View workflow job for this annotation

GitHub Actions / pre-commit

Ruff (E501)

vllm/entrypoints/openai/serving_chat.py:718:81: E501 Line too long (85 > 80)
choice_data = ChatCompletionResponseChoice(
index=output.index,
message=message,
Expand Down
4 changes: 3 additions & 1 deletion vllm/transformers_utils/tokenizers/mistral.py
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@
Tekkenizer)

from vllm.logger import init_logger
from vllm.utils import is_list_of
from vllm.utils import is_list_of, generate_valid_mistral_tool_id

if TYPE_CHECKING:
from vllm.entrypoints.chat_utils import ChatCompletionMessageParam
Expand Down Expand Up @@ -61,7 +61,9 @@
while True:
try:
tool_call = next(tool_calls_validator) # type: ignore
tool_call['id'] = generate_valid_mistral_tool_id()
logger.warning(f"Assigned new tool_id: {tool_call['id']} for tool: {tool_call}")
validated_tool_calls.append(tool_call)

Check failure on line 66 in vllm/transformers_utils/tokenizers/mistral.py

View workflow job for this annotation

GitHub Actions / pre-commit

Ruff (G004)

vllm/transformers_utils/tokenizers/mistral.py:66:25: G004 Logging statement uses f-string

Check failure on line 66 in vllm/transformers_utils/tokenizers/mistral.py

View workflow job for this annotation

GitHub Actions / pre-commit

Ruff (E501)

vllm/transformers_utils/tokenizers/mistral.py:66:81: E501 Line too long (88 > 80)
except StopIteration:
break

Expand Down
10 changes: 10 additions & 0 deletions vllm/utils.py
Original file line number Diff line number Diff line change
Expand Up @@ -52,11 +52,16 @@
import vllm.envs as envs
from vllm.logger import enable_trace_function_call, init_logger

from random import choices
from string import ascii_letters, digits

if TYPE_CHECKING:
from vllm.config import VllmConfig

logger = init_logger(__name__)

ALPHANUMERIC = ascii_letters + digits

# Exception strings for non-implemented encoder/decoder scenarios

# Reminder: Please update docs/source/features/compatibility_matrix.md
Expand Down Expand Up @@ -2206,3 +2211,8 @@ def run_method(obj: Any, method: Union[str, bytes, Callable], args: Tuple[Any],
else:
func = partial(method, obj) # type: ignore
return func(*args, **kwargs)

def generate_valid_mistral_tool_id():
# Mistral Tool Call Ids must be alphanumeric with a maximum length of 9.
# https://github.com/mistralai/mistral-common/blob/21ee9f6cee3441e9bb1e6ed2d10173f90bd9b94b/src/mistral_common/protocol/instruct/validator.py#L299
return "".join(choices(ALPHANUMERIC, k=9))
Loading