The JC dashcam series and VL trackers put video next to position data. The platform treats an ADAS or driver-monitoring event as a first-class record with its footage attached, not as an alert you then go hunting for.
Video evidence attached to the event that triggered it, which is what makes a driver-behaviour programme survive contact with a disputed incident.
Protocol-level facts from the manufacturer’s own documentation. Model-level behaviour varies within a series, so confirm your exact unit with us rather than assuming.
| Item | Supported | Notes |
|---|---|---|
| Protocol family | JT/T 808 derivatives, GT06-family variants | Varies by series — confirm per model |
| Transport | TCP for telemetry; media on its own channel | Media bandwidth is sized separately |
| Series commonly deployed | JC (dashcam), VL (vehicle tracker), LL (asset) | Video only on the JC range |
| ADAS events | Forward collision, lane departure, headway | Raised by the device, stored with footage |
| DSM events | Fatigue, distraction, phone use, smoking | Driver-facing camera required |
| Media retrieval | On-demand clip pull and scheduled upload | Full-time upload is rarely worth the data cost |
| Storage | On-device SD plus platform-side retention | Retention length drives hosting size |
Most platform vendors publish a list of supported model numbers. It looks reassuring and it is usually out of date, because it describes a snapshot of what somebody tested once rather than how the integration actually works. GPS platforms do not implement models; they implement protocols. Jimi IoT devices report over JT/T 808 derivatives and GT06-family variants across TCP for telemetry, separate channel for media, and the platform decodes that protocol — which is why a new unit in the JC (dashcam), VL (vehicle), LL (asset) families generally works on the day it ships rather than after a development cycle.
What genuinely varies between models is not the protocol but the hardware: which inputs physically exist, whether a CAN interface is present, whether there is a relay output, how many 1-Wire devices can be attached. That is why we ask for your exact model and firmware version rather than publishing a matrix. It takes us a day to confirm against a live server, and it is the difference between a deployment that works and one that discovers a missing input after the units are installed.
Send the model, the firmware version and what you need to report on. You will get back a straight answer on three things: whether the unit reports into the platform at all, which of your required data points it can actually produce, and whether a cheaper unit in the same range would do the same job. That last one costs us margin and saves you money, which is the point — a fleet that buys the wrong hardware blames the platform.
Jimi IoT units already in the field can be re-pointed to a new server remotely, so a migration does not mean visiting vehicles. We audit what is connected first, import your historical data, re-point a pilot group, and cut the rest across only once the pilot reconciles cleanly against your old system. The old endpoint stays live through the window. Full detail is on the hosting and migration page.
Send your Jimi IoT model list and we will confirm compatibility against a live server before you commit. Request a compatibility check or message us on WhatsApp — include firmware versions if you have them, and tell us what you need to report on.
It should not, and on any real deployment it does not. Continuous upload is expensive in data and rarely useful — nobody watches it. The platform pulls clips on demand and accepts event-triggered uploads, so the footage you store is the footage attached to something that actually happened.
ADAS watches the road: forward collision warning, lane departure, headway. DSM watches the driver: fatigue, distraction, phone use. ADAS needs a forward-facing camera; DSM needs a driver-facing one. Most fleets start with ADAS because it is less contentious with drivers, and add DSM once the programme has credibility.
Enough that it changes your hosting sizing, which is why we ask about it before quoting. Position data for a thousand vehicles is predictable; video depends entirely on event rate and retention period. Tell us how long you must keep footage — often set by an insurer or a client contract — and it sizes properly.
Not automatically, and a programme that ignores this fails. What works in practice is introducing ADAS first, sharing the scoring openly rather than using it punitively, and being specific about what is recorded and who can see it. The technology is the easy part of a DSM rollout.
Yes, and most fleets do — video on the vehicles where an incident is expensive, basic tracking everywhere else. Both report into the same platform, and driver scoring works across the whole fleet with video evidence available on the subset that carries it.
Send us your exact models and firmware versions. We will test them against a live server and tell you what works before you commit to anything.
Check My Device →