Model catalog

The right model for the job.

Quantum models share one API contract, so you can tune speed, context, and reasoning without rewriting your integration.

Fast

Free

quantum-1.0

Use in the API →

Quantum 1.0 is a capable general-purpose model designed for production workloads. It handles a wide range of tasks including text generation, classification, extraction, summarization, and structured output. Use it as your default for assistants, content pipelines, and applications that need reliable, consistent responses.

Context window

128k

Best for

Everyday production

Use cases

  • Chat assistants
  • Content generation
  • Data extraction
  • Summarization
  • Q&A systems

Very fast

Free

quantum-1.0-flash

Use in the API →

Quantum 1.0 Flash is optimized for speed and efficiency. It delivers rapid responses for lightweight tasks where latency matters more than depth. Ideal for classification, quick answers, and high-throughput workloads where you need fast turnarounds at scale.

Context window

128k

Best for

High-volume, low-latency tasks

Use cases

  • Classification
  • Quick questions
  • High-volume extraction
  • Lightweight chat
  • Rapid prototyping

Premium

Pro

quantum-1.5

Use in the API →

Quantum 1.5 is our most capable model, offering expanded context windows and richer reasoning. It excels at complex tasks that require deeper understanding, multi-step reasoning, and handling of nuanced instructions. Available exclusively on the Pro plan.

Context window

256k

Best for

Complex reasoning and demanding tasks

Use cases

  • Complex reasoning
  • Long-form content
  • Advanced coding
  • Agent workflows
  • Detailed analysis

Choosing a model

Start with quantum-1.0 if you're unsure. It's a strong default for most production workloads and everyday AI tasks.

Use quantum-1.0-flash when you need fast responses for lightweight tasks like classification, quick answers, or high-volume extraction.

Upgrade to quantum-1.5 for complex reasoning, long-form content, advanced coding, and tasks that benefit from a larger context window.