xAI · Released Mar 9, 2026Grok 4.20 (Reasoning)
Reasoning Grok for document-heavy analysis and long-horizon tool use.
- Capability
- 152.0ECI · #41 of 148
- Input
- $1.25per 1M tokens
- Output
- $2.50per 1M tokens
- Context
- 1M30K max output
What it costs
- Input
- $1.25per million tokens
- Output
- $2.50per million tokens
- Cached input
- $0.20per million tokens
- Over 200K tokens
- $2.50 / $5.00input / output per million
- Blended (3:1)
- $1.56pricier than 65% of priced models
Typical monthly bills
| Monthly usage | Estimated cost |
|---|---|
| Side project2M input + 0.5M output tokens | $3.75 |
| Team assistant25M input + 5M output tokens | $43.75 |
| Production app250M input + 50M output tokens | $437.50 |
Official xAI API price, as listed on models.dev.
Independent benchmarks
#41 of 148 scored models
The shaded band is Epoch AI’s confidence range (149.3–154.4); the tick marks the median scored model.
Scores from Epoch AI, run independently of xAI.
The details
- Lab
- xAI
- Released
- Mar 9, 2026
- Context window
- 1,000,000 tokens
- Max output
- 30,000 tokens
- Inputs
- Text, Images, PDFs
- Output
- Text
- Reasoning
- Always on
- Tool calling
- Yes
- Structured output
- Yes
- Weights
- Proprietary
- API model ID
grok-4.20-0309-reasoningon xAI- Availability
- 11 API providerslisted on models.dev
More from xAI
About Grok 4.20 (Reasoning)
How much does Grok 4.20 (Reasoning) cost?
Grok 4.20 (Reasoning) costs $1.25 input / $2.50 output per million tokens (official xAI API price). Cached input is $0.20 per million tokens. Requests over 200K tokens are billed at $2.50 input / $5.00 output. At a 3:1 input-to-output mix that is $1.56 per million tokens, more expensive than 65% of the 360 priced models we track.
What is the context window of Grok 4.20 (Reasoning)?
Grok 4.20 (Reasoning) accepts up to 1,000,000 tokens per request and can write up to 30,000 tokens in one response.
How good is Grok 4.20 (Reasoning)?
Epoch AI gives Grok 4.20 (Reasoning) a Capabilities Index score of 152.0 (likely range 149.3–154.4), ranking it #41 of 148 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 89.3%, FrontierMath Tiers 1–3 44.9%, OTIS Mock AIME 2024–2025 92.2%, SimpleQA Verified 30.2%.
Is Grok 4.20 (Reasoning) open source?
No. Grok 4.20 (Reasoning) is proprietary; you use it through xAI’s API or partner platforms.
What inputs does Grok 4.20 (Reasoning) support?
Grok 4.20 (Reasoning) accepts text, images and PDFs and replies in text. It is a reasoning model, supports tool calling and can return structured JSON output.
When was Grok 4.20 (Reasoning) released?
xAI released Grok 4.20 (Reasoning) on Mar 9, 2026.
What are the best alternatives to Grok 4.20 (Reasoning)?
The closest current models from other labs on capability, price and release date are Kimi K2.6 (Moonshot AI, ECI 151.1, $0.95 / $4.00), Gemini 3 Flash Preview (Google, ECI 151.8, $0.50 / $3.00), GLM-5.2 (Z.ai (Zhipu), ECI 151.8, $1.40 / $4.40) and GPT-5.4 mini (OpenAI, ECI 148.8, $0.75 / $4.50).