Skip to content

AI that moveswork forward.

One OpenAI-compatible endpoint routes every request across Lite, Pro, or Max — from real-time speed to million-token reasoning.

Start building↗See the platform↗

One request enters.

The application sends one familiar request. DMS reads what the work demands before a model ever answers.

The right intelligence answers.

Lite for latency. Pro for precision. Max for context. One response contract comes back to your product.

  1. InputOpenAI-compatible
  2. ProfileLatency · precision · context
  3. Selectedpro-256k
  4. OutputStreamed response
One endpointdms-auto-routerLite64K · latencyPro256K · precisionMax1M · contextProtocolOpenAI compatibleResponseStreaming + tools

One model ID. Four decisions.

dms-auto-router

The endpoint profiles every request by latency, precision, and context before it selects a tier. Your client keeps one response contract from first token to final output.

Receive

Accept one OpenAI-compatible request.

POST /v1/chat/completions

Profile

Read the latency, precision, and context signals.

workload.profile

Route

Select the strongest available tier for the work.

dms-auto-router

Stream

Return one consistent response shape.

200 OK · SSE

The workload changes.
The endpoint does not.

Lite, Pro, and Max are operating profiles—not a list of model names. Each one is tuned around the constraint that matters most to the work.

01lite-64k

Speed & volume

Lite

Real-Time FastLow-latency inference for customer-facing chat, high-volume classification, and real-time AI agents.Explore Lite
64Ktoken context window
Real workloadResolve a live customer request.Latency first
Input
Support history + current message
Route
lite-64k
Output
Low-latency streamed answer
02pro-256kMost popular

Code & precision

Pro

Code & PrecisionStructured generation accuracy, multi-file coherence, and instruction fidelity for precision-sensitive workloads.Explore Pro
256Ktoken context window
Real workloadRefactor a multi-file service.Precision first
Input
Repository context + instructions
Route
pro-256k
Output
Structured patch + tool calls
03max-1mFlagship

Reasoning & context

Max

Flagship ReasoningLong-context reasoning, multi-document synthesis, and multi-step agent orchestration at scale.Explore Max
1Mtoken context window
Real workloadAnalyze an entire codebase.Context first
Input
Full repository + extended history
Route
max-1m
Output
Cross-system synthesis and plan
ROUTE

The request finds the right intelligence.

One OpenAI-compatible endpoint routes each workload to the model tier built for it.

Model selection and streaming
GROUND

Private context arrives before the answer.

Your knowledge layer grounds the request without turning private data into training material.

Retrieval and context assembly
ACT

Reasoning becomes an action.

Tool-aware models choose the next operation, return structured output, and continue the workflow.

Tool use and orchestration
VERIFY

Every outcome keeps an operational trail.

Policies, deployment boundaries, and audit signals keep production workloads observable.

Policy and audit controls

Proof the architecture, not the promise.

The application keeps a familiar interface while routing, context, tools, and deployment controls stay visible underneath.

/v1/chat/completions

One endpoint

Keep the SDK your team already knows. Change the base URL and model ID.

64K · 256K · 1M

Purpose-built tiers

Move from fast conversations to coding and long-context reasoning without rebuilding the integration.

Shared · Dedicated · Private

Deployment control

Start on shared infrastructure, then scope dedicated or private deployment with the DMS team.

Keep the SDK. Change the infrastructure.

The integration stays familiar while your team gains access to DMS model tiers and production deployment options.

OPENAI SDKapi.dmslab.ai
import OpenAI from "openai";

const dms = new OpenAI({
  apiKey: process.env.DMSLAB_API_KEY,
  baseURL: "https://api.dmslab.ai/v1",
});

const response = await dms.chat.completions.create({
  model: "dms-auto-router",
  messages: [
    { role: "user", content: "Review this production change." }
  ],
  stream: true,
});
RESPONSE TRACE
LITE

A customer gets an answer while the moment is live.

route = "lite-64k" · stream = true

Latency-prioritized routing for chat, classification, and interactive agents.

Deploy where trust requires it.

Start with the managed API. Move toward reserved or private infrastructure when your workload and controls demand it.

Three inference tiers. One path to private.

Choose the workload profile that fits today. Prices stay synced with the existing DMS API; dedicated and private infrastructure remain a separate enterprise path.

Lite

lite-64k · Real-Time Fast

Low-latency inference for customer-facing chat and real-time agents.

$2.50/wk64K
  • Customer-facing chat and support
  • High-volume classification and routing
  • Latency-prioritized model selection
Choose plan

Pro

pro-256k · Code & Precision

Structured generation accuracy and multi-file coherence for precision-sensitive work.

$6/wk256K
  • Multi-file generation and refactoring
  • Instruction-heavy structured outputs
  • Benchmark-led model selection
Choose plan

Max

max-1m · Flagship Reasoning

Long-context reasoning and multi-step agent orchestration at scale.

$25/wk1M
  • Full-codebase and document analysis
  • Extended-history agent workflows
  • Availability shown in dashboard
Choose plan

Build the first request. Scope the real system.

Use the shared API today, or work with DMS Lab on capacity, network boundaries, and deployment architecture.

Start building Talk to infrastructure