#Japan #Tokyo #AI Gateway #FlashLabs #OrcaRouter

FlashLabs Launches OrcaRouter for Cost-Effective AI Solutions

On June 3, 2026, FlashLabs Inc., located in Chiyoda, Tokyo, announced the launch of its OrcaRouter subscription plan, specifically designed to address the rising costs associated with AI while ensuring high-quality output akin to flagship models. This innovative solution allows users to access over 200 LLM models, such as Claude Opus 4.8, GPT-5.5 Pro, and Gemini 3.5, with up to a 10% bonus credit.

Background and Objectives

By 2026, the enterprise AI market is projected to undergo rapid expansion, with cost efficiency becoming an increasingly pressing issue. The Japanese AI market is expected to grow from $7.9 billion in 2025 to a staggering $39.1 billion by 2034, exhibiting an annual growth rate of 18.80%. As a result, companies are hastening their investments in AI. However, deploying every request to expensive frontier models like Claude Opus and GPT-5.5 Pro can lead to excessive costs. The alternative—adapting routing to accommodate new models—often results in obsolescence, placing the financial and maintenance burden on development teams.

With 37% of enterprises using five or more models in production by 2026, the AI routing market is shifting from simply substituting cheaper models to intelligently selecting the optimal model on a prompt basis. OrcaRouter has been developed as a next-generation AI gateway that addresses this challenge by evaluating the quality and routing in real-time.

What is OrcaRouter?

OrcaRouter boasts several groundbreaking features:

• - Adaptive Routing: Determines the difficulty of prompts and automatically routes complex requests to frontier models while simpler tasks are sent to open models.
• - Support for Over 200 Models: Includes popular APIs such as Claude Opus 4.8, GPT-5.5 Pro, and Gemini 3.5, among others.
• - LinUCB Context Bandit: Learns from request outcomes and minimizes the allocation to underperforming models.
• - Low Latency: Routing decisions are made in less than 1ms, ensuring a seamless user experience.
• - Comprehensive Transparency: Each decision’s outcome, model, provider, and publicly listed price are recorded on a per-request basis for full visibility.

Pricing

• - Monthly Plan: Offers automatic bonus credits of up to 10% for commitments made each billing cycle.
• - Token Billings: No additional charges over the public price set by the provider, ensuring transparency (0% markup).
• - Zero Routing Fees: No additional costs associated with routing decisions.

Value Proposition for Businesses

1. Significant Cost Savings: OrcaRouter can cut down AI costs by approximately 40% while maintaining output quality. By routing standard tasks—like extraction, classification, formatting, and simple summarization— to open models at about 1/15 of the usual costs, only advanced reasoning tasks utilize frontier models. This results in a projected annual savings of around $47,700 for teams spending roughly $10,000 monthly. The payback period is less than a day!
2. Transparent Pricing: With 0% added fees on tokens and routing, and the potential for bonus credits, companies can maximize their return on AI investments.
3. Easy Integration: With an OpenAI-compatible API, integration is as simple as changing a base URL in existing code. This facilitates low-cost implementation and seamless transitions between testing and production using established workflows with tools like Cursor, Cline, and LangChain.

Technical Features

OrcaRouter intelligently connects to multiple providers like Anthropic and OpenAI, applying their respective terms directly without middlemen, which increases flexibility in service selection according to internal policies.

• - Learning-Based Bandit Approach: Instead of basic if/else logic, the model learns from performance, decreasing allocation to models that perform poorly within specified prompts.
• - Full Visibility: Every decision made by OrcaRouter, from model selection to provider pricing, is documented and available via headers or dashboards, allowing complete auditability.

Security and Compliance

OrcaRouter integrates eight guardrail features to uphold enterprise-level security and compliance, enhancing operational control for production environments. These include safeguards against personal information leakage, API key protection, and prompt injection attempts, ensuring safe operational practices.

Future Plans

FlashLabs aims to accelerate the adoption of AI among Japanese businesses through OrcaRouter, enhancing it with stronger guardrail functions and continuous support for high availability (99.99% SLA). The company will also expand Japanese-language documentation and explore custom offerings to fit the unique needs of the Japanese market.

Statement from Leadership

Yōichi Hosoi, CEO of FlashLabs, stated, “OrcaRouter represents a fresh paradigm of intelligent routing that addresses the rising costs associated with AI usage. By empowering businesses to select the optimal model per prompt, we allow for the retention of high-quality performance while slashing costs by about 40%. Our transparent pricing model will enable firms to maximize their AI investments, fostering innovation and competitive advantage in the global market.”

About OrcaRouter

OrcaRouter is developed by Continuum AI, based in the United States, and exclusively distributed by FlashLabs in Japan. It routes requests based on prompt difficulty, ensuring high-grade performance while reducing costs significantly. Its low-latency mechanisms and extensive model library are tailored to meet enterprise demands efficiently.

About FlashLabs

FlashLabs aims to automate and eventually autonomize sales and customer experiences through AI. With its rich portfolio, including OrcaRouter, FlashIntel, and Chroma, it is positioned to deliver exceptional results by integrating machine speed with human insight.

For further inquiries, please contact:
FlashLabs Inc. Marketing Department
Email: [email protected]
Official Website: FlashLabs