easydel.inference.tools.parsers.mistral_tool_parser

easydel.inference.tools.parsers.mistral_tool_parser#

class easydel.inference.tools.parsers.mistral_tool_parser.MistralToolCall(*, id: str = <factory>, type: str = 'function', function: ~easydel.inference.openai_api_modules.FunctionCall, **extra_data: ~typing.Any)[source]#

Bases: ToolCall

static generate_random_id()[source]#
id: str#
static is_valid_id(id: str) → bool[source]#
model_config: ClassVar[ConfigDict] = {'extra': 'allow'}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class easydel.inference.tools.parsers.mistral_tool_parser.MistralToolParser(tokenizer: AutoTokenizer)[source]#

Bases: ToolParser

Tool call parser for Mistral models (7B Instruct v0.3+).

Designed for use with: - [mistral_common](mistralai/mistral-common) - the examples/tool_chat_template_mistral.jinja template

Handles Mistral’s specific tool call format with [TOOL_CALLS] token and JSON array of function calls. Supports both standard JSON parsing and regex-based parsing for function names with v11+ tokenizers.

Features: - Automatic tool ID generation (9-character alphanumeric) - Support for multiple tool calls in single response - Streaming with incremental argument parsing - Compatibility with different Mistral tokenizer versions

Used when –enable-auto-tool-choice –tool-call-parser mistral are set.

adjust_request(request: ChatCompletionRequest) → ChatCompletionRequest[source]#

Adjust request parameters for model-specific requirements.

Override this method to modify request parameters like system prompts, tool definitions, or formatting before processing. Default implementation returns the request unchanged.

Parameters

request – Original chat completion request

Returns

Potentially modified request

Return type

ChatCompletionRequest

Example

Some models may need to reformat tool definitions or add specific system instructions for proper tool calling.

extract_tool_calls(model_output: str, request: ChatCompletionRequest) → ExtractedToolCallInformation[source]#

Extract tool calls from complete Mistral model response.

Parses the [TOOL_CALLS] token followed by JSON array or function call patterns. Handles both standard JSON format and regex-based extraction for newer tokenizer versions.

Parameters

  • model_output – Complete model output with tool calls

  • request – Original request (unused)

Returns

Extracted tool information with MistralToolCall objects

Note

Tool call arguments should avoid quotes as parser may need to replace single quotes with double quotes.

extract_tool_calls_streaming(previous_text: str, current_text: str, delta_text: str, previous_token_ids: Sequence[int], current_token_ids: Sequence[int], delta_token_ids: Sequence[int], request: ChatCompletionRequest) → easydel.inference.openai_api_modules.DeltaMessage | None[source]#

Extract tool calls from streaming model output.

Processes incremental model output to identify partial tool calls and emit appropriate streaming updates. Maintains state across calls to handle incomplete JSON/XML structures.

Parameters

  • previous_text – Text accumulated up to previous call

  • current_text – Text accumulated including current chunk

  • delta_text – New text in current chunk

  • previous_token_ids – Token IDs up to previous call

  • current_token_ids – Token IDs including current chunk

  • delta_token_ids – New token IDs in current chunk

  • request – Original request with tool definitions

Returns

Incremental tool call update, or None if no update

Return type

DeltaMessage

Raises

NotImplementedError – Must be implemented by subclasses

Note

This method is stateful - it uses instance variables to track parsing progress across streaming chunks.

Contents