Deploy AI translation engines, machine translation post editing, and translation quality assurance within Smartcat’s translation management system to minimize manual review and maintain consistent quality.
Pick from powerful LLMs (GPT-4o, GPT-5, Claude) or custom MT models tuned for your domain and project type. All Custom.MT processing happens in secure infrastructure.
By combining automated translation + automatic post-editing + quality gates, you get end-to-end AI localization that cuts down on manual work and accelerates time to market.
Why Use Custom.MT in Smartcat
Efficiency: Translate and post-edit automatically — less manual labor, fewer cost leaks.
Proven performance: Combine Smartcat’s industry-trusted TMS with Custom.MT’s GenAI power.
Scalability: Run high-volume workflows without bottlenecking your team.
Full customization: Set up prompts, models, and rules to match your style and quality needs.
How It Works — Smartcat + Custom.MT Integration
In Smartcat's project settings or translation engine settings, configure a custom translation provider (Custom.MT) via API / webhooks.
Set up credentials or API key in your Custom.MT Console.
In Custom.MT Console, go to Connections → Smartcat.
Choose which models to use (MT, APE, QE) for each language pair.
Define QE thresholds, post-editing prompts, and terminology rules.
When you trigger translation from Smartcat, it sends segments to Custom.MT.
Custom.MT returns translated (and optionally post-edited / quality scored) segments to Smartcat.
High-confidence segments can be auto-approved; lower-confidence ones flagged for review.
Reviewed or approved segments live in Smartcat Editor.
You export the finished translation or sync it directly back to your system.
Example Use Cases
See how teams leverage Custom.MT with Smartcat across different content types
Product UI / Mobile App Localization
Send UI strings from Smartcat to Custom.MT, run APE and MTQE then review only problematic segments.
Marketing content for global campaigns
Translate landing pages or ad copy in bulk, enforce brand messaging via custom prompts.
Technical / Help Center Content
Use QE to catch low-quality MT, then apply post-editing to critical segments.
Simple, Transparent Pricing
Choose the perfect plan for your localization needs
For small translation teams and LSP
Note:
The usage of machine translation APIs and GenAI tokens is charged on top of the platform fees. Pricing.
Teams that localize over 10 million target words per year
The usage of machine translation APIs and GenAI tokens is charged on top of the platform fees. Pricing.
Large businesses with compliance requirements
The usage of machine translation APIs and GenAI tokens is charged on top of the platform fees. Pricing.
Ready to transform your localization workflow?
Start using Custom.MT with Smartcat today and experience the power of AI-driven translation
No, you just add Custom.MT as a custom translation provider via API or callback, then configure it in your Custom.MT Console.
Yes, in the Custom.MT Console, you define model settings, APE prompts, and QE thresholds per language pair or project type.
You can use QE to flag or reroute low-scoring segments to Smartcat editors, or run a dual-pass workflow (QE → APE → QE).
You can enforce terms via Custom.MT prompts or import your Smartcat glossaries for use in APE to maintain brand consistency.
Yes. Custom.MT processes translation data in secure infrastructure, does not retain client data long-term, and supports enterprise-grade authentication.
Yes. In the Custom.MT Console, you can assign different models or engines per language pair or content category. Smartcat will always call Custom.MT, and Custom.MT decides which engine to use for each segment
This usually happens when:
You can view processing logs inside the Custom.MT Console under Translations → Logs to see the exact reason.
Custom.MT processes content at the segment level, matching Smartcat’s default segmentation. Document-level context can be simulated using advanced prompts or custom workflows.