entropic::Gemma4Adapter Class Reference

Gemma 4 chat adapter (covers A4B / E4B / E2B variants). More...

#include </home/runner/work/entropic/entropic/src/inference/adapters/gemma4_adapter.h>

Inheritance diagram for entropic::Gemma4Adapter:

Collaboration diagram for entropic::Gemma4Adapter:

Public Member Functions
std::string	chat_format () const override
	Chat format: GGUF-embedded template.
ParseResult	parse_tool_calls (const std::string &content) const override
	Parse tool calls via tagged JSON, falling back to bare JSON.
	ChatAdapter (std::string tier_name, std::string identity_prompt)
	Construct adapter with tier identity.
Public Member Functions inherited from entropic::ChatAdapter
	ChatAdapter (std::string tier_name, std::string identity_prompt)
	Construct adapter with tier identity.
std::string	format_system_prompt (const std::string &base_prompt, const std::vector< std::string > &tool_jsons) const
	Assemble system prompt: identity + context + tools.
virtual Message	format_tool_result (const ToolCall &tool_call, const std::string &result) const
	Format a tool result as a user message.
virtual std::string	format_tools (const std::vector< std::string > &tool_jsons) const
	Format tool definitions for injection into system prompt.
bool	is_response_complete (const std::string &content, const std::vector< ToolCall > &tool_calls) const
	Check if response represents task completion.
virtual std::string	format_system_with_vision (const std::string &base_system, bool has_vision) const
	Format system prompt with optional vision context.
virtual std::string	format_content_parts (const std::vector< ContentPart > &parts) const
	Convert multimodal content parts to adapter-specific format.

Additional Inherited Members
Protected Member Functions inherited from entropic::ChatAdapter
std::vector< ToolCall >	parse_tagged_tool_calls (const std::string &content) const
	Parse <tool_call>JSON</tool_call> tagged blocks.
std::vector< ToolCall >	parse_bare_json_tool_calls (const std::string &content) const
	Parse bare JSON lines containing "name" key.
std::string	extract_thinking (const std::string &content) const
	Extract <think>...</think> content.
std::string	strip_think_blocks (const std::string &content) const
	Strip all <think>...</think> blocks from content.
std::optional< ToolCall >	try_recover_json (const std::string &json_str) const
	Attempt JSON recovery on malformed tool call string.
std::optional< ToolCall >	parse_single_tool_call (const std::string &json_str) const
	Parse a single JSON tool call string.
Protected Attributes inherited from entropic::ChatAdapter
std::string	tier_name_
	Identity tier name.
std::string	identity_prompt_
	Assembled identity prompt.
std::unordered_set< std::string >	tool_prefixes_
	Known tool prefixes.

Detailed Description

Gemma 4 chat adapter (covers A4B / E4B / E2B variants).

Uses the GGUF-embedded chat template via chat_format()="" and a permissive multi-format tool-call parser pending empirical verification.

Version

2.1.9

Definition at line 62 of file gemma4_adapter.h.

Member Function Documentation

chat_format()

std::string entropic::Gemma4Adapter::chat_format ( ) const

inlineoverridevirtual

Chat format: GGUF-embedded template.

Returns

Empty string — llama.cpp applies the stored template. @utility

Version

2.1.9

Implements entropic::ChatAdapter.

Definition at line 72 of file gemma4_adapter.h.

ChatAdapter()

entropic::ChatAdapter::ChatAdapter	(	std::string	tier_name,
		std::string	identity_prompt
	)

Construct adapter with tier identity.

Parameters

tier_name	Identity tier (e.g. "eng", "lead").
identity_prompt	Assembled identity prompt.

Version

1.8.2

Parameters

tier_name	Identity tier name.
identity_prompt	Assembled identity prompt.

Version

1.8.2

Definition at line 66 of file adapter_base.cpp.

parse_tool_calls()

ParseResult entropic::Gemma4Adapter::parse_tool_calls ( const std::string &  content ) const

overridevirtual

Parse tool calls via tagged JSON, falling back to bare JSON.

Parse tool calls from Gemma 4 output.

Permissive strategy pending empirical confirmation of Gemma 4's native tool-call syntax. Strips <think> blocks from content.

Parameters

content

Raw model output.

Returns

ParseResult.

Version

2.1.9

Tries tagged JSON first (<tool_call>{...}</tool_call> and the asymmetric <|tool_call> / <|im_start|>tool_call channel variants the base parser now accepts), then bare-JSON lines as a fallback. The base class handles malformed JSON recovery transparently when either path is exercised.

Parameters

content

Raw model output.

Returns

ParseResult with cleaned content and any extracted calls.

Implements entropic::ChatAdapter.

Definition at line 35 of file gemma4_adapter.cpp.

---

The documentation for this class was generated from the following files:

• inference/adapters/gemma4_adapter.h
• inference/adapters/gemma4_adapter.cpp