# Grok 4.20 (Reasoning) API prices

Grok 4.20 (Reasoning) (grok) — 2.0M context, 2.0M max output.
15 metered per-token offers from 12 provider(s), in USD per 1M tokens.
Prices as of 2026-10-04.

| Provider | Input $/1M | Output $/1M | Cache read $/1M | Context |
| --- | --- | --- | --- | --- |
| Eden AI | $1.25 | $2.5 | $0.2 | 1.0M |
| Vertex | $1.25 | $2.5 | $0.2 | 2.0M |
| Impossibl | $1.25 | $2.5 | $0.2 | 1.0M |
| DevPass (LLM Gateway) | $1.25 | $2.5 | $0.2 | 2.0M |
| LLM Gateway | $1.25 | $2.5 | $0.2 | 2.0M |
| OCI Generative AI | $1.25 | $2.5 | $0.2 | 1.0M |
| Tempr Gateway | $1.25 | $2.5 | $0.2 | 1.0M |
| Vercel AI Gateway | $1.25 | $2.5 | $0.2 | 2.0M |
| Vercel AI Gateway | $1.25 | $2.5 | $0.2 | 2.0M |
| xAI | $1.25 | $2.5 | $0.2 | 1.0M |
| Azure | $2 | $6 | — | 262k |
| Cloudflare AI Gateway | $2 | $6 | $0.2 | 2.0M |
| DevPass (LLM Gateway) | $2 | $6 | $0.2 | 2.0M |
| LLM Gateway | $2 | $6 | $0.2 | 2.0M |
| Ofox | $4 | $12 | $0.4 | 2.0M |

## Summary

- Cheapest input: $1.25 per 1M tokens (Eden AI)
- Cheapest output: $2.5 per 1M tokens (Eden AI)
- First-party: $2.5 per 1M output tokens (xAI)
- Free offers: none
- Subscription plans (not per-token): none
- Released: 2026-03-09
- Inputs: image, pdf, text

## FAQ

### What is the cheapest Grok 4.20 (Reasoning) API?

As of Oct 4, 2026, Eden AI has the lowest Grok 4.20 (Reasoning) output price at $2.5 per 1M tokens, and Eden AI has the lowest input price at $1.25 per 1M tokens. Prices are metered per-token rates in USD; free tiers and subscription plans are excluded.

### How much does Grok 4.20 (Reasoning) cost on xAI?

xAI charges $1.25 per 1M input tokens and $2.5 per 1M output tokens for Grok 4.20 (Reasoning).

### How many providers offer Grok 4.20 (Reasoning)?

12 providers list Grok 4.20 (Reasoning) on Sovyron; 15 of them sell it at a metered per-token price.

### Is Grok 4.20 (Reasoning) free?

No provider in the Sovyron catalog lists a free tier for Grok 4.20 (Reasoning).

### What is the context window of Grok 4.20 (Reasoning)?

Grok 4.20 (Reasoning) supports a 2.0M-token context window and up to 2.0M output tokens.

HTML page: https://sovyron.com/models/grok-4-20-reasoning/
Full dataset: https://sovyron.com/data/catalog.json