Skip to content
View in the app

A better way to browse. Learn more.

Картографф

A full-screen app on your home screen with push notifications, badges and more.

To install this app on iOS and iPadOS
  1. Tap the Share icon in Safari
  2. Scroll the menu and tap Add to Home Screen.
  3. Tap Add in the top-right corner.
To install this app on Android
  1. Tap the 3-dot menu (⋮) in the top-right corner of the browser.
  2. Tap Add to Home screen or Install app.
  3. Confirm by tapping Install.

1 Screenshot

078 — Sgki

In short, SGKI 078 is marketed as a that can run sophisticated neural‑network models locally, without draining a battery.

SGKI 078 isn’t just another sensor; it’s a that blends low‑power compute, robust security, and flexible connectivity into a single, ultra‑compact package. While the ecosystem is still maturing, the early developer response suggests a strong appetite for on‑device intelligence that can operate for months without recharging.

The SDK also supports and FreeRTOS , giving developers flexibility in choosing a lightweight OS or a bare‑metal approach. sgki 078

Key high‑level claims from SGI’s pre‑release brief:

If you’re planning a product that needs , keeping an eye on SGKI 078 (or ordering a prototype kit) is a smart move. Expect the first wave of commercial devices—smart wearables, industrial monitors, and low‑cost security nodes—to hit the market by late 2026. In short, SGKI 078 is marketed as a

SGI has released , a free, open‑source software suite that includes:

Grab the SGKI 078‑EVB, dive into the SGK‑SDK, and start turning raw sensor streams into actionable intelligence—right at the edge. The SDK also supports and FreeRTOS , giving

A standout feature is the system. The chip can shut down the NPU, sensor hub, or even the RISC‑V core independently, based on workload. Coupled with an ultra‑low‑leakage SRAM (2 KB for model weights, 4 KB for runtime buffers), the module can stay in “deep sleep” for months on a coin‑cell battery.

| Feature | Detail | |---------|--------| | | 12 mm × 12 mm × 2.5 mm (M.2‑compatible) | | Power envelope | ≤ 1 mW typical, sub‑µW standby | | Processing | Integrated 64‑bit RISC‑V core @ 250 MHz with on‑chip NPU | | AI capabilities | 4 TOPS (tera‑operations per second) for inference | | Connectivity | BLE 5.2, Thread, optional LoRaWAN module | | Security | On‑chip secure element, hardware root of trust, post‑quantum ready crypto |

B2B marketplaces often list these parts with full diagnostic data, including load capacity, heat resistance, and material composition. Why Codes Like SGKI 078 Matter

Account

Navigation

Search

Search

Configure browser push notifications

In short, SGKI 078 is marketed as a that can run sophisticated neural‑network models locally, without draining a battery.

SGKI 078 isn’t just another sensor; it’s a that blends low‑power compute, robust security, and flexible connectivity into a single, ultra‑compact package. While the ecosystem is still maturing, the early developer response suggests a strong appetite for on‑device intelligence that can operate for months without recharging.

The SDK also supports and FreeRTOS , giving developers flexibility in choosing a lightweight OS or a bare‑metal approach.

Key high‑level claims from SGI’s pre‑release brief:

If you’re planning a product that needs , keeping an eye on SGKI 078 (or ordering a prototype kit) is a smart move. Expect the first wave of commercial devices—smart wearables, industrial monitors, and low‑cost security nodes—to hit the market by late 2026.

SGI has released , a free, open‑source software suite that includes:

Grab the SGKI 078‑EVB, dive into the SGK‑SDK, and start turning raw sensor streams into actionable intelligence—right at the edge.

A standout feature is the system. The chip can shut down the NPU, sensor hub, or even the RISC‑V core independently, based on workload. Coupled with an ultra‑low‑leakage SRAM (2 KB for model weights, 4 KB for runtime buffers), the module can stay in “deep sleep” for months on a coin‑cell battery.

| Feature | Detail | |---------|--------| | | 12 mm × 12 mm × 2.5 mm (M.2‑compatible) | | Power envelope | ≤ 1 mW typical, sub‑µW standby | | Processing | Integrated 64‑bit RISC‑V core @ 250 MHz with on‑chip NPU | | AI capabilities | 4 TOPS (tera‑operations per second) for inference | | Connectivity | BLE 5.2, Thread, optional LoRaWAN module | | Security | On‑chip secure element, hardware root of trust, post‑quantum ready crypto |

B2B marketplaces often list these parts with full diagnostic data, including load capacity, heat resistance, and material composition. Why Codes Like SGKI 078 Matter