All AI models · LLM benchmarks · Methodology

OrcaRouter Fusion

Curated fan-out router that runs Claude Opus 4.8, GPT-5.5 and Gemini 3.1 Pro in parallel on hard requests, then has a Claude Opus 4.8 judge return the strongest single answer verbatim. Easy requests fall through to a cheaper default and bill as one call.

Model facts

Context window
1M tokens
Maximum output
128K tokens
Cheapest paid input
not published per 1M tokens
Cheapest paid output
not published per 1M tokens
Open weights
no
Providers
1

Capabilities and modalities

reasoning, tool calling, structured output, attachments, text, image, pdf.

Observed price history

  • 2026-08-28: not published input / not published output per 1M tokens

API providers

  • OrcaRouter — model id orcarouter/fusion; not published input / not published output per 1M tokens; provider documentation