// AI ROUTING INFRASTRUCTURE

AI MODEL
VAULT.

MVS is the AI routing layer for the entire MelTuc platform. Every blueprint that needs an LLM sends its request to MVS, which dispatches to the active provider — OpenRouter, Ollama, DeepSeek, or MiniMax — with automatic fallback. One API. Every model. Zero provider lock-in.

Open Dashboard Learn More
4
Providers
1
API Endpoint
<50ms
Dispatch Overhead
Auto
Fallback Routing

Live Request Monitor

The dashboard shows the active routing configuration alongside real-time request activity. Today's request count and total token consumption appear at the top. The recent requests table shows the last ten dispatches with app ID, provider name, success status, latency, and timestamp. A top-apps strip highlights which blueprints are making the most calls today.

meltuc.tech/mvs/app
1,247
Requests Today
2.4M
Tokens Today
99.2%
Success Rate
342ms
Avg Latency
ACTIVE CONFIG
OpenRouter → anthropic/claude-haiku-4.5
fallback: Ollama / gemma4:26b
RECENT REQUESTS
App ID Provider Status Latency Time
SRMOpenRouterok1,234ms14:22:01
AIFOpenRouterok987ms14:21:44
PFGOllamaok412ms14:21:30
MVS_TESTDeepSeekerr5,001ms14:20:55

Enable and Manage Providers

The providers page shows every registered provider and all their associated models. Each provider and model can be individually enabled or disabled with a toggle. Model identifiers are the exact strings passed to the provider's API. Context window sizes are shown for planning prompt budgets.

meltuc.tech/mvs/providers
OpenRouter enabled
OpenRouterAdapter
anthropic/claude-haiku-4.5 · 200k ctx
anthropic/claude-sonnet-4-6 · 200k ctx
meta-llama/llama-3.3-70b · 128k ctx
Ollama (Local) enabled
OllamaAdapter
gemma4:26b · 8k ctx
DeepSeek disabled
DeepSeekAdapter
deepseek-v4-flash · 1M ctx

Active Config & Token Management

The settings page is where you set the active provider and model, configure an optional fallback for when the primary fails, and set default temperature and max token limits. The token section lets you generate per-app bearer tokens for external integrations. Each token is identified by app ID and can be individually enabled or disabled without regenerating.

meltuc.tech/mvs/settings
ACTIVE CONFIGURATION
Primary Provider
OpenRouter
Primary Model
anthropic/claude-haiku-4.5
Fallback Provider
Ollama (Local)
Temperature
0.7
Max Tokens
2000
ACTIVATE CONFIG
API TOKENS
SRMenabled
AIFenabled
PFGdisabled
+ Generate New Token

Live Chat API Tester

The test interface lets you fire a live request through the full dispatch pipeline without writing any code. Enter a system prompt and user message, optionally override the provider, and submit. The response displays with timing, token usage, and the full normalised output. Every test run is saved to the history view for later reference.

meltuc.tech/mvs/test
SYSTEM PROMPT
You are a helpful assistant...
USER MESSAGE
Summarise the Anthropic MCP spec in 3 bullet points.
Use active config
SEND
RESPONSE
• MCP defines a standard protocol for connecting AI models to external tools and data sources
• It uses a server/client model where the LLM acts as a client
• Version 2 introduces breaking changes to the message envelope format
Provider: OpenRouter Latency: 1,247ms Tokens: 312
// SUPPORTED PROVIDERS

Four Adapters, One API

MVS ships with adapters for four providers. The dispatch layer normalises all responses to the same output schema regardless of which adapter handled the call. Switching the active provider requires no code changes in any calling blueprint.

OpenRouter
Cloud API gateway with access to hundreds of models including all Anthropic, Meta, and Mistral offerings
cloud200k+ ctx
Ollama (Local)
Local inference on the VPS GPU. Zero latency variability, no API costs, full privacy for sensitive data
localfree tier
DeepSeek
DeepSeek V4 Flash and V4 Pro (Reasoning) via the DeepSeek API. 1M context. Competitive on cost
cloud1M ctx
MiniMax
MiniMax M2.5 with 1M context window for long-document and multi-session workloads
cloud1M ctx
// WHAT YOU GET

The Full AI Routing Stack in One App

MVS is the single integration point for all AI across the MelTuc platform — every blueprint that needs an LLM routes through here.

🔀

Provider-Agnostic Routing

One dispatch call works with OpenRouter, Ollama, DeepSeek, or MiniMax. Switch the active provider in settings — zero code changes required in any calling blueprint.

Automatic Fallback

Configure a primary and fallback provider. When the primary fails or times out, MVS retries against the fallback transparently — callers never see the error.

📈

Live Request Monitor

Today's request count, token consumption, success rate, and avg latency updated on every page load. The recent requests table shows the last ten dispatches with full metadata.

🔑

Per-App API Tokens

Generate bearer tokens scoped to individual blueprints. Each token can be individually enabled or disabled without regenerating. External integrations authenticate via header.

🧪

Live Test Interface

Fire real requests through the full dispatch pipeline from the browser — no code, no curl. System prompt, user message, optional provider override. Response, timing, and token usage all returned.

🔌

FMR Model Sync

Free models discovered and validated by FMR are synced directly into MVS with a single admin action — provisioned, skipped, and error counts reported for complete visibility.

🛟

Zero-Config Dispatch

Platform code routes through MVS with a single call_llm(prompt, via_mvs=True) — no token, no router import. The internal dispatch endpoint accepts a flat body and falls back to a direct provider call if MVS is ever unreachable.

💰

Unified Cost Attribution

Every successful call is logged to both the MVS ledger and the platform-wide AI operations log — source app, model, token counts, and cost — feeding ecosystem-wide usage and spend reporting.

// GET STARTED

One API. Every Model.

Route all platform AI traffic through a single, configurable dispatch layer with automatic fallback. No provider lock-in. No code changes when switching models.

Open Model Vault

Requires a MelTuc account. Create one free.