TFLite(LiteRT) - Generative AI on Genio 420

This page lists the generative AI models and representative performance data for Genio 420 platforms. For background information about generative workloads and usage notes, refer to TFLite(LiteRT) - Generative AI.

Model Support and Performance

Note

Performance numbers apply to both Android and Yocto (this platform supports both OS). For OS-level GAI support across all platforms, see AI Supporting Scope.

The following symbols are used in the tables below:

  • -- : To be released.

  • Q3/E : Support planned for Q3 (estimated).

  • X : Platform does not support this model.

Large Language Models (LLMs)

Performance Comparison (Prompt/Generative) (Unit: tok/s)

Model

Genio 420

Qwen3-0.6B

–

Qwen3-1.7B

218.85 / 12.04

Qwen3-4B

–

Qwen3-8B

–

Qwen2.5-1.5B-Instruct

291.13 / 17.92

Qwen2.5-3B-Instruct

126.80 / 9.45

Qwen2.5-7B-Instruct

–

gemma3-1B (Text Only)

492.98 / 23.94

gemma3-4B (Text-Only)

–

gemma2-2b-it

X

llama3.2-1B-Instruct

331.70 / 22.55

llama3.2-3B-Instruct

121.42 / 9.21

llama3-8b

49.26 / 4.30

MiniCPM-2B-sft-bf16-llama-format

–

Phi-3-mini-4k-instruct

–

Phi-3.5-mini-instruct

–

DeepSeek-R1-Distill-Qwen-1.5B

–

DeepSeek-R1-Distill-Qwen-7B

X

DeepSeek-R1-Distill-Llama-8B

X

Android-only Runtime Models

The following models use Android-only speculative decoding runtime and are not available on Yocto.

Performance Comparison (Prompt/Generative) (Unit: tok/s)

Model

Genio 420

llava1.5-7b-speculative-decoding

X

medusa_v1_0_vicuna_7b_v1.5

X

vicuna1.5-7b-tree-speculative-decoding-plus

X

baichuan-7b-int8-cache

X

Vision-Language Models (VLMs)

Performance Comparison (ViT Inference Time / Prompt / Generative)

Model

ViT Inference Time (s)

Prompt Mode (tok/s)

Generative Mode (tok/s)

Qwen3VL-2B

–

–

–

InternVL3-1B

–

–

–

Stable Diffusion and Image Generation

Performance Comparison (Main Time/Inference Time) (Unit: ms)

Model

Genio 420

Stable Diffusion v2.1 base model with controlnet

X

Stable Diffusion v.1.5 controlnet

–

CLIP and Embedding Models

Performance Comparison (Main Time/Inference Time) (Unit: ms)

Model

Genio 420

img_encoder_proj_clip_vit_large_dynamic

–

img_encoder_proj_openclip_vit_big_g_dynamic

X

img_encoder_proj_openclip_vit_h_dynamic

X

text_encoder_clip_vit_large

–

text_encoder_openclip_vit_h

X

Speech Recognition Models

Support Status

Model

Genio 420

Whisper

Q3/E