Product Launchllmon device inferenceedge computingtiiny ai
Tiiny AI Launches Pocket Lab Running 120B LLM
10.0
Tiiny AI Inc. unveiled the Tiiny AI Pocket Lab in Hong Kong, verified by Guinness World Records and shown ahead of CES January 2026. The pocket-sized device claims to run up to a 120-billion-parameter LLM fully on-device, within a 65W typical power envelope and delivering ≈190 TOPS with 80GB LPDDR5X and 1TB SSD. It supports one-click deployment of major open-source models and offline private workflows.
Key Points
- 1Unveils pocket-sized device running up to 120B-parameter LLM entirely on-device, offline.
- 2Reduces energy and privacy costs by executing large models locally within a 65W typical power envelope.
- 3Enables developers and professionals to run long-context, multi-step agent workflows without cloud dependency.
Scoring Rationale
Official product unveiling demonstrates unprecedented on-device 120B LLM capability, high potential but currently relies on vendor benchmarks and claims.
Sources
Public references used for this report.
Practice with real Logistics & Shipping data
90 SQL & Python problems · 15 industry datasets
Used by DS/ML engineers at top companies
High-Value Overnight OrdersEasyDelivered International ShipmentsMediumOn-Time Delivery Rate by CarrierHard
250 free problems · No credit card
See all Logistics & Shipping problems
