Conheça o TrueForge: o agent harness de código aberto e independente de fornecedor. Custo 50% menor. Explorar agora→

Novo | Registro de Habilidades para agentes de IA

Auto Routing: send every request to the optimal model that can handle it

Stop paying frontier prices for routine queries. Optimize model spend by matching each request to the model it actually needs.

Agende uma Demonstração

Leia a documentação

Purple gradient square with white background, shiny surface, and rounded corners in rhombus shape.

Proven cost savings

Book a Demo

69%

Cheaper

Measured against an all-Opus 5 baseline.

98%

Quality

at 3.2x cheaper per correct answer.

1.9x

Faster

Mean latency fell from 7.6 seconds to 4.0

Habilidades de Agente

Reduza o uso de tokens com carregamento de contexto sob demanda

Assign a model to the simple, medium, and complex tiers behind one virtual model name, and the gateway picks the tier per request. Swapping in a cheaper model next quarter is a config edit rather than a release.

Ler Mais

Skills registry dashboard with search and list of skills, repositories, and descriptions displayed.

Classification Strategies

Salve e reutilize a mesma Habilidade em vários agentes

Heuristic classification uses out-of-the-box parameters to instantly classify prompts, adding zero cost or latency. LLM classification calls a small model you nominate to accurately identify prompt complexity.

Ler Mais

Skills registry repository creation form with fields for repository, skill name, description, and skill instructions.

Conversation Pinning

Empacote mais do que apenas texto de prompt

Once a tier answers a turn successfully, pinning keeps later turns of that conversation at the same tier or higher. A short follow-up cannot drop an active conversation onto a cheaper model.

Ler Mais

Code editor displaying SKILL.md file with default skill template and structured instructions and capabilities listed.

Pronto para Empresas

Implante um gateway de IA seguro que mantém seus dados e modelos dentro da sua infraestrutura em nuvem / on-premise.

Conformidade e Segurança
Padrões SOC 2, HIPAA e GDPR para garantir uma proteção de dados robusta
Governança e Controle de Acesso
SSO + Controle de Acesso Baseado em Função (RBAC) e Registro de Auditoria
Suporte Empresarial e Confiabilidade
Suporte 24/7 com garantia de SLA SLAs de resposta

Implante o TrueFoundry em qualquer ambiente

VPC, no local, isolado ou em várias nuvens.

Nenhum dado sai do seu domínio. Desfrute de soberania completa, isolamento e conformidade de nível empresarial onde quer que o TrueFoundry seja executado

Começar

Cloud deployment options including On-Prem, Multi-Cloud, Air-gapped, and AWS, Google Cloud Platform.

O AI Gateway da TrueFoundry padronizou a forma como cada equipe interage com LLMs, embeddings e componentes RAG. Em vez de integrações dispersas, agora controlamos o acesso, as políticas de roteamento e as barreiras de segurança centralizadamente. A capacidade de otimizar custos ou latência sem alterar as aplicações tem sido um divisor de águas. Isso tornou nossa arquitetura de IA mais limpa, mais segura e muito mais fácil de escalar.

Smiling man with beard and short dark hair in a collared shirt and blazer.

Indroneel G.

Diretor de TI, Siemens Healthineers

Perguntas frequentes

O que é uma Habilidade no Skills Registry?

Auto Routing is a routing strategy in TrueFoundry's AI Gateway that classifies each incoming request as simple, medium, or complex and sends it to the model you assigned to that tier. Unlike weight, priority, or latency-based routing, the decision depends on the content of the request itself.

Como as Habilidades TrueFoundry são diferentes dos prompts de sistema?

In our benchmarks, Auto Routing was 69% cheaper than an all-Opus 5 setup on graded academic datasets and 80% cheaper on production-shaped traffic. Route away from a prior-generation top model instead and graded savings land closer to 50%. The gateway logs the resolved tier for every request, so you can measure your own savings directly.

Como um Registro de Habilidades difere de um Registro de Servidor MCP Local?

Quality retention was 98% of the baseline under deterministic grading. The known limit is that the free heuristic classifier reads the shape of a prompt rather than its true difficulty, so a short but genuinely hard question can be routed to a cheap tier. Switching that virtual model to the LLM classifier recovers most of that quality for roughly ten points less in savings.

Que problemas o TrueFoundry Skills Registry resolve?

The heuristic classifier scores the prompt against a fixed set of signals inside the gateway process. It is free, fully deterministic, and adds no latency. The LLM classifier calls a small model you nominate to judge each request, which is more accurate on ambiguous prompts but adds one billable model call and some latency per request.

Como o TrueFoundry Skills Registry reduz o uso de tokens?

No. Auto Routing is a setting on a virtual model, and your application keeps calling that virtual model's name. Changing tiers, swapping models, or switching classification strategy happens in configuration.

Como as Skills são criadas, compartilhadas e gerenciadas no TrueFoundry?

It can move up and it will not move down. Conversation pinning records the tier that successfully served a turn and keeps later turns at that tier or higher. Pins are automatic, need no configuration, and last through ten minutes of inactivity.

What are the current limitations?

Auto Routing supports exactly three tiers, and the heuristic classifier's signals are fixed rather than customizable. It runs on virtual models only, for the chat, completion, and responses model types. The dashboard form supports one target per tier; multiple targets in a tier require the YAML manifest or the API.