Skip to content

Products

GPU CloudAMD MI325X and NVIDIA H200 by the hour or on reservation.
Dedicated ClustersPreviewPrivate, interconnected GPU clusters for sustained workloads.
SandboxesPreviewIsolated, short-lived environments for developers and AI agents.
Serverless GPUPreviewGPU containers that scale with requests, down to zero.
Serverless InferenceOpen models behind one OpenAI-compatible endpoint.
Dedicated EndpointsReserved serving capacity for one model, billed per GPU-hour.
PlaygroundSend a prompt to any model your workspace can reach.
Model DeploymentsPreviewImmutable deployments with instant promote and rollback.
Fine-tuningLoRA fine-tunes on the open models in our catalog.
Model TrainingPreviewFoundation-model training runs, managed end to end.
EvaluationsPreviewBenchmark suites and model comparisons you can rerun.
DatasetsManaged and synthetic training data, versioned and ready to use.
Data ManagementPreviewDatasets, connections and storage, with lineage across the platform.
PipelinesPreviewVersioned DAGs that automate data, training and deployment.
Data RoomsPreviewSecure document sharing with guests, NDAs and an audit trail.
NokuvaPreviewThe AI-native design editor.
TavocPreviewA Rust-native code platform: IDE and desktop apps.
FoltracComing soonMulti-cloud deployment with autonomous operations.
W0rkTreeComing soonVersion control for agent fleets.
Identity & AccessMembers, roles, policies, SSO and audit for every workspace.
Billing & CostUsage, invoices, credits and budgets in one place.
Logs & MonitoringRequest logs, latency and error rates across services.
Project CanvasPreviewSee and change a project's resources on one live canvas.

Pages

Home/
Platform/platform
All products/products
Pricing/pricing
GPU pricing/pricing#compute
Inference services/services/inference
Model training services/services/model-training
Protocol design/services/protocol-design
Research partnerships/services/research-partnerships
Protocols/protocols
Datasets/datasets
Documentationdocs.legionedge.ai
Changelog/changelog
News/news
Research/research
Blog/blog
Articles/articles
Events/events
Startup Program/programs/startup
Ambassador Program/programs/ambassador
Partner Program/programs/partner
Event Program/programs/events
Company/company
Careers/careers
Partners/partners
Contact/contact
Statusstatus.legionedge.ai
Terms of Service/legal/terms
Privacy Policy/legal/privacy

Context layer

v1.5Stable

Model Context Protocol

Structured context interfaces for optimal model comprehension.

Available for
  • Rust
  • TypeScript
  • Python
cargo add legion-mcp
model-context-protocol.tsTypeScript
import { server, resource } from "@legionedge/mcp";
 
const docs = resource("docs", { schema: DocSchema });
 
server({ resources: [docs] })
  .tool("search", SearchInput, async (q) => index.query(q))
  .listen();
Version
v1.5
Status
Stable
License
Open specification
Implementations
Rust · TypeScript · Python
Stack layer
Context

What the spec defines

The guarantees, not the vibes.

  1. Typed resources — context arrives as schema-validated data, not string glue

  2. Tool definitions declared once, not re-sent every turn

  3. Prompt templates as data, versioned alongside the spec

  4. Conformance-tested across Rust, TypeScript, and Python

Overview

Why it's built this way.

Spec, schemas and conformance suite

The Model Context Protocol defines structured interfaces for feeding context to language models: typed resources, tool definitions, and prompt templates that models comprehend reliably, replacing the ad-hoc string concatenation that most applications still ship.

Structure pays twice. Models read validated, predictable context more reliably — and applications stop re-sending boilerplate every turn. Our field study of production MCP traffic found tool-definition overhead and redundant re-sends were two of the three largest sources of context waste; declare-once semantics remove both.

Tavoc's context assembly is built on the protocol end to end: scoped repository views arrive as typed resources, tools are declared at session start, and every context decision is inspectable in the session log.

Implementations are available for Rust, TypeScript, and Python, all conformance-tested against the same spec suite. The spec and the field-study data behind its design decisions are in the research registry.

Proven in production

If a protocol can't survive contact with a real product, it isn't finished. This one ships inside Tavoc.

Explore Tavoc

Implement it

Implementing the Model Context Protocol?

The conformance suite is public, and the team that wrote the spec reads its issue tracker. Independent implementations are the point.