Grok 4.20 Multi-Agent pricing, context window and limits
Grok 4.20 Multi-Agent costs $1.25 per million input tokens and $2.5 per million output tokens at xAI Grok list price. It accepts up to 1,000,000 tokens of context and returns up to 30,000 tokens.
- Model id
grok-4.20-multi-agent-0309 - Snapshot
What does Grok 4.20 Multi-Agent cost?
| Input | $1.25 |
|---|---|
| Output | $2.5 |
| Cached input (read) | $0.2 |
| Cache write | — |
US dollars per million tokens. A dash means no price is published.
How much can Grok 4.20 Multi-Agent read and write?
| Context window | 1,000,000 |
|---|---|
| Max output | 30,000 |
What can Grok 4.20 Multi-Agent do?
| Vision (image input) | Yes |
|---|---|
| Tool calling | No |
| Structured output | Yes |
| Reasoning | Yes |
| Input | text, image, pdf |
| Output | text |
What does a request to Grok 4.20 Multi-Agent cost?
| A typical request: 1,000 tokens in, 500 out | $0.0025 |
|---|---|
| A long request: 10,000 tokens in, 1,000 out | $0.015 |
| 1,000 typical requests | $2.5 |
| 1 million tokens in and 1 million out | $3.75 |
Input read from the provider's prompt cache costs $0.2 per million tokens instead of $1.25, 84% less. A request that repeats a 1,000-token prompt from the cache costs $0.0014 instead of $0.0025.
Among 8 xAI Grok models with a published price, Grok 4.20 Multi-Agent is the 2nd cheapest by input price. Its context window is the largest of 13. List prices only; discounts, batch prices and prices above 200K tokens are not shown.
How do I call Grok 4.20 Multi-Agent through DVARA?
Configure the grok provider in the DVARA LLM Gateway and send the model id grok-4.20-multi-agent-0309. DVARA, the AI governance platform, applies your policies, audit and cost attribution to the call. See LLM cost attribution and budgets.
In DVARA, if you haven't set your own price for a model, its calls are costed at the price in the catalogue that ships with your DVARA version. A price you set in Flightdeck always wins. You can switch catalogue prices off. Then a model with no price you set is counted as costing nothing. Every cost record says which price it used.
What should I check before relying on this data?
The weekly review marked this entry. A person saw each mark and kept the model listed.
- The provider charges more above 200K tokens of input; the table shows the base price.
Which other xAI Grok models are there?
- Grok 4.20 (Reasoning)$1.25 in · $2.5 out · 1M context
- Grok 4.3$1.25 in · $2.5 out · 1M context
- Grok 4.20 (Non-Reasoning)$1.25 in · $2.5 out · 1M context
- Grok 4.5$2 in · $6 out · 500K context
- Grok 4.6$2 in · $6 out · 500K context
- Grok 4.7$2 in · $6 out · 500K context
- Grok Build 0.1$1 in · $2 out · 256K context
- Grok Imagine Image— in · — out · 16K context
- Grok Imagine Image Quality— in · — out · 16K context
- Grok Imagine Video— in · — out · 1K context
- Grok Imagine Video 1.5— in · — out · 1K context
- Grok Imagine Video 1.5 Lite— in · — out · 1K context
Which models from other providers are similar to Grok 4.20 Multi-Agent?
These are the closest in context window and input price, one per provider.
- Grok 4.3Amazon Bedrock · $1.25 in · $2.5 out · 1M context
- Gemini 2.5 ProGoogle Gemini · $1.25 in · $10 out · 1.05M context
- GLM-5.2Zhipu GLM · $1.4 in · $4.4 out · 1M context
- GLM-5.2Mistral AI · $1.4 in · $4.4 out · 1M context
- Qwen3.8 MaxAlibaba Qwen · $1.7774 in · $5.3323 out · 1M context