All AI models · LLM benchmarks · Methodology

grok-4.20-multi-agent-beta-0309

Grok model for agentic tool use, reasoning, coding, and live assistance

Model facts

Context window
2M tokens
Maximum output
3K tokens
Cheapest paid input
$2 per 1M tokens
Cheapest paid output
$6 per 1M tokens
Open weights
no
Providers
2

Capabilities and modalities

reasoning, tool calling, attachments, temperature control, text, image.

Observed price history

  • 2026-08-24: $2 input / $6 output per 1M tokens

API providers

  • 302.AI — model id grok-4.20-multi-agent-beta-0309; $2 input / $6 output per 1M tokens; provider documentation
  • Arena — model id grok-4.20-multi-agent-beta-0309; not published input / not published output per 1M tokens; provider documentation