Does OpenVINO Model Server have a free tier?
Open-source software (Apache-2.0) with no API bill at any point; the cost is your own machine's compute and electricity.
OpenVINO Model Server: get an API key and write fiction with it on your phone
The Intel route for people without an NVIDIA card: run int4-quantized models straight on CPU, iGPU, or Arc GPUs, with a ready-made pre-converted model library on Hugging Face — even an office laptop can host a private continuation backend. This page is the BYOK quick answer for OpenVINO Model Server: key steps, direct console and pricing links, and the app-side notes that matter. Verified 2026-07-18.
Two links up front: get your key at the official download page (it runs locally — there is no pricing page because there is no bill). Prices and free quotas move; the official pages are authoritative.
What does it power inside Foreverse?
The factory preset declares 1 modality flag for OpenVINO Model Server. The models actually available come from the live catalog fetched in-app — catalogs move fast, and the flag only means the protocol wiring is ready.
| Modality | Used for |
|---|---|
| Text | Continuation · roleplay chat · agent |
How do I install OpenVINO Model Server and reach it from my phone?
- Install Docker on an Intel machine (CPU-only works; for iGPU/Arc acceleration use the latest-gpu image and pass --device /dev/dri).
- One command starts the server and auto-downloads a pre-converted model: docker run -d -p 8000:8000 -v $(pwd)/models:/models openvino/model_server:latest --source_model OpenVINO/Qwen3-8B-int4-ov --model_repository_path /models --task text_generation --rest_port 8000.
- Use pre-converted int4 weights from the OpenVINO organization on Hugging Face (like OpenVINO/Qwen3-8B-int4-ov) — no format conversion on your side.
- Allow port 8000 through the firewall; with phone and computer on the same Wi-Fi, set the app host to http://<computer-IP>:8000/v3.
Key in hand: three steps to connect
- Open the app's Settings → AI models & services and find OpenVINO Model Server in the list (one of 62 presets, with logo and search).
- Paste the key and tap the connectivity test — it fires one real request to verify key and network, and pulls back the model list while it's at it.
- Enable the models you want, then set per-modality defaults (text, image, speech, and video each get their own); continuation, chat, and listening route through them from then on.
The one universal host pitfall: the API host must include its full version prefix (this provider's exact form is in the notes below) — the app never appends /v1 for you.
OpenVINO Model Server-specific notes
The preset host is http://localhost:8000/v3 — note the /v3 version prefix, not /v1 (OVMS mounts its OpenAI-compatible endpoints under /v3). When swapping localhost for the machine's LAN/Tailscale IP, keep the full /v3: entering just ip:port makes the app append a generic /v1, which is wrong here. No auth, so a placeholder key works; the model ID must match the name loaded on the server (like OpenVINO/Qwen3-8B-int4-ov).
FAQ
Where does the key live once I paste it into the app?
It's stored on your device only, under system-level encryption — never uploaded or synced to any Foreverse server. Requests go straight from your phone to the provider's official endpoint with zero markup. To revoke access, delete the key in the provider's console.