For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).

Python class

GrammarEnforcementSnapshot

GrammarEnforcementSnapshot

class max.pipelines.context.GrammarEnforcementSnapshot(in_thinking_region, grammar_enforced, tool_calling_match_buffer, thinking_match_buffer)

source

Bases: object

Captured grammar-enforcement state for rollback.

The speculative bitmask path walks the enforcement state through draft tokens to compute downstream slot constraints and then restores this snapshot so committed-token processing on the next batch replays the same transitions from a clean state. Lives next to TextContext because it exposes TextContext.snapshot_grammar_state() / TextContext.restore_grammar_state(); the concrete implementation in TextContext constructs and consumes instances.

Parameters:

• in_thinking_region (bool)
• grammar_enforced (bool)
• tool_calling_match_buffer (list[int])
• thinking_match_buffer (list[int])

grammar_enforced

grammar_enforced: bool

source

in_thinking_region

in_thinking_region: bool

source

thinking_match_buffer

thinking_match_buffer: list[int]

source

tool_calling_match_buffer

tool_calling_match_buffer: list[int]

source

Was this page helpful?

Thank you! We'll create more content like this.

Thank you for helping us improve!

😔 What went wrong?

Some code doesn’t work

It includes inaccurate information

It's missing information I need

It was difficult to understand

Other

Submit