Overview & Capability
Google ultra-compact, on-device foundation model engineered for low-latency, privacy-first mobile, IoT, and edge execution across Pixel, Android, and Chrome.
Key Highlights
- Local multimodal inference with zero cloud latency
- On-device summarization, proofreading, and smart replies
- Native hardware acceleration via NPUs and GPUs
Robbie's Independent Take
Verified: 2026-10-11
Excellent candidate for local execution with zero cloud lock-in. For 7B-8B parameter models, budget at least 8 GB of unified memory or VRAM (Q4_K_M quantization). For 32B-70B models, aim for 24 GB+ to maintain interactive throughput without CPU offload penalties.
Alternatives & Replaces
- Local quantized Llama models
- Phi-3 Mini