Skip to main content

Hal 1.0

Hal 1.0 is the highest capability tier. It handles complex reasoning, multi-step agents and deep code analysis. Use Hal when the task pays for quality: research synthesis, architecture decisions, long-form generation and agent loops that decompose problems into subtasks.

Model string

Specs

Pricing is dynamic. For current rates see radium.cloud/pricing.

Capabilities

  • Streaming: Supported
  • Tool calling: Supported
  • JSON mode: Supported
  • JSON schema: Supported
  • Reasoning: Supported
  • Vision input: Supported
  • Multi-step agent reasoning
  • Deep code review and architecture analysis
  • Research synthesis across large document sets
  • Complex structured output generation
  • Chain-of-thought reasoning workflows

Not suited for

High-volume classification, real-time chat, or any workload where latency and cost are the primary constraints. Use Clarke 1.0 or Tycho 1.0 for those.

Example request