REFFT AI Store

GLM-4.6V-FlashMeasured

Instruction-tuned multimodal GLM-4.6V-Flash language model for on-device text generation.

4.50tok/s

TPS / Decode

Text-to-TextQualcomm Hexagon NPUNVIDIA GPUAMD GPUApple NPUw4QCS9075

LFM2.5-2.6BMeasured

LiquidAI LFM2.5-2.6B hybrid dense decoder for on-device text generation.

Text-to-TextQualcomm Hexagon NPUw4SM8850+1

LFM2.5-8B-A1BMeasured

LiquidAI LFM2.5-8B-A1B hybrid MoE language model for on-device text generation.

Text-to-TextQualcomm Hexagon NPUw4QCS9075

LFM2.5-VL-3BMeasured

LiquidAI LFM2.5-VL-3B hybrid decoder + SigLIP2 (text-first Hexagon pack).

Image-Text-to-TextQualcomm Hexagon NPUw4SM8850+1

Qwen3-0.6BMeasured

Qwen3-0.6B dense decoder language model for on-device text generation.

Text-to-TextQualcomm Hexagon NPUw4QCS9075

Qwen3-1.7BMeasured

Qwen3-1.7B dense decoder language model for on-device text generation.

Text-to-TextQualcomm Hexagon NPUw4QCS9075

Qwen3-4B-Instruct-2507Measured

Qwen3-4B-Instruct-2507 dense decoder language model for on-device text generation.

Text-to-TextQualcomm Hexagon NPUw4QCS9075

SmolVLM2-500M-Video-InstructMeasured

Compact vision-language model that captions images and answers questions about them on-device, with text, image and video-frame input.

Image-Text-to-TextQualcomm Hexagon NPUNVIDIA GPUAMD GPUApple NPUw4SM8850+1

A native serving and inference tool.