FfiInferenceModelSession class
FFI implementation of InferenceModelSession. Buffers query chunks until getResponse is called.
Routes all per-conversation native calls through handle — its own
ConversationHandle — so multiple sessions on one model are fully
isolated. extractTextFromResponse is a static helper on
LiteRtLmFfiClient and needs no instance.
- Inheritance
-
- Object
- InferenceModelSession
- FfiInferenceModelSession
- Mixed-in types
Constructors
- FfiInferenceModelSession({required ConversationHandle handle, required ModelType modelType, required ModelFileType fileType, required bool supportImage, required bool supportAudio, bool enableThinking = false, required VoidCallback onClose})
Properties
- enableThinking → bool
-
final
- fileType → ModelFileType
-
final
- handle → ConversationHandle
-
final
- hashCode → int
-
The hash code for this object.
no setterinherited
- lastRawResponse → String?
-
Most recent raw SDK JSON. Returns the response of the last getResponse
or getResponseAsync. For Gemma 4 use
LiteRtLmFfiClient.extractToolCallson this string to surface tool calls.no setteroverride - modelType → ModelType
-
final
- onClose → VoidCallback
-
final
- runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
- supportAudio → bool
-
final
- supportImage → bool
-
final
Methods
-
addQueryChunk(
Message message) → Future< void> -
override
-
close(
) → Future< void> -
override
-
getResponse(
) → Future< String> -
override
-
getResponseAsync(
) → Stream< String> -
override
-
getSessionMetrics(
) → SessionMetrics -
Get session metrics including token usage and performance stats.
override
-
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
sizeInTokens(
String text) → Future< int> -
override
-
stopGeneration(
) → Future< void> -
override
-
toString(
) → String -
A string representation of this object.
inherited
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited