Grok 4.20 Multi-Agent
xAI · Multimodal · Released Mar 2026
Grok 4.20 Multi-Agent is xAI's extended-context model supporting multi-agent orchestration and long-context reasoning, with a 2-million-token context window.
- Strengths
- Handles extremely long contexts and multi-agent workflows that require coordinating multiple reasoning threads or processing massive documents and codebases.
- Best for
- Tasks requiring agents to collaborate on complex problems, processing full repositories or books in a single request, and long-running reasoning chains.
- Limitations
- Predated by Grok 4.3, which offers frontier capabilities with a 1M token window; use Grok 4.7 for the latest agentic and software engineering workloads at scale.
Input / 1M
$1.25
Output / 1M
$2.50
Cached input / 1M
$0.2
Context window
2M
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in — out —
- 90d
- in — out —
- 1y
- in decreased 37.5% out decreased 58.3%
- Since launch
- in decreased 37.5% out decreased 58.3%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 16 Jun 2026 | $1.25 | $2.50 | $0.2 | Imported from OpenRouter | openrouter.ai |
| 11 Jun 2026 | $2.00 | $6.00 | $0.2 | Imported from OpenRouter | openrouter.ai |