local_ai_gemma 0.0.3

flutter_gemma adapter for LocalAI Kit: implements the core LocalLlm interface on top of Google's on-device Gemma runtime.

Changelog

0.0.3

• Forward per-request temperature, top-p and maximum-token settings to the native inference session.
• Apply recommended sampling defaults for Qwen 3.5 models.

0.0.2

• Type the flutter_gemma runtime boundary for stronger compile-time guarantees.
• Coordinated version bump; correct the bundled LICENSE text.

0.0.1

• Initial public release.