OpenRouter Logo

OpenRouter

Unified Aggregator Intelligent Routing Live Benchmarks

Unified model aggregation with intelligent routing and competitive pricing.

The OpenRouter AI Provider (uno.anahata.asi.openrouter.OpenRouterAiProvider) connects your Anahata ASI container to openrouter.ai. Through a single API key, gain instant, native access to hundreds of frontier and open-source models from OpenAI, Anthropic, Google, Meta, DeepSeek, Mistral, and xAI (Grok).

Deep Metadata Discovery

Anahata's specialized OpenRouterModel automatically queries OpenRouter's /v1/models endpoint and discovers the true architectural specifications of every available model:

  • Input Context Lengths: Maps context_length and top_provider.context_length directly to maxInputTokens (from 8K up to 1M+).
  • True Output Modalities: Reads architecture.output_modalities cleanly, eliminating ambiguity between input perception and output response generation.
  • Native Function Calling: Automatically inspects supported_parameters for tools, enabling Anahata's Java execution and workspace tools without manual toggling.
  • Sampling Defaults: Pre-seeds optimal temperature, top-p, and top-k values directly from the model creator's default_parameters.

Provider-Specific Output Ceilings

Unlike generic OpenAI proxies that assume a static 4K or 16K output ceiling, OpenRouter reports the top provider's maximum single-turn completion capacity:

For long-generation models like Grok 4.6, top_provider.max_completion_tokens reveals up to 450,000 output tokens. Anahata ASI automatically captures this limit into maxOutputTokens and dynamically negotiates generation headroom during deep coding and architectural turns.

Live Benchmarks & Pricing Transparency

OpenRouter bundles live performance and economics directly into its model responses:

Anahata ASI extracts per-token prompt and completion costs and embeds them cleanly into model descriptions (e.g. [In: $2.00/M | Out: $6.00/M]). Live Artificial Analysis coding and intelligence benchmarks are rendered directly inside model tooltips and inspection cards.

1-Click Configuration

Getting started with OpenRouter in Anahata ASI takes seconds:

  1. Obtain an API key from openrouter.ai/keys.
  2. Open the AI Providers settings panel in Anahata ASI (NetBeans, Desktop, or IntelliJ).
  3. Select OpenRouter from the provider list and paste your API key into the Key Pool.
  4. Click Test Connection to instantly discover hundreds of models and start prompting!