Which Grok model has the largest context window?

Grok 4.3 and the listed 4.20 variants have 1M token context windows, larger than the Grok 4.5 flagship at 500k. Grok Build 0.1 has a 256k context window.

Which Grok model has the lowest listed input price?

Grok Build 0.1 has the lower listed input price in the current data. Check official docs and console access before relying on it.

Do these prices include image or video?

No. This table covers text model API token pricing. Other features can have separate pricing.

Is the cheapest model always the best default?

No. Reliability, context size, output quality and review time can matter more than the listed token price.

How often should I re-check model names?

Re-check before launch, before large budget changes and whenever the official xAI docs announce new or retired model IDs.

Grok Models Compared: Specs and Prices

Model comparison is useful only when it is tied to a job. The table below shows the current model data used by this site, but you should still test with your own prompts, files, tools and output requirements. A model that is cheaper per token can still be more expensive if it needs more retries or produces answers that take longer to review.

If you are choosing a chat subscription rather than an API model, use the plan chooser or pricing page instead. API model names matter most when your code sends requests directly to xAI.

Model	Context	Input per 1M	Output per 1M	Best for
Grok 4.5 grok-4.5	500k	USD 2.00	USD 6.00	Current flagship for general chat, coding and reasoning
Grok 4.3 grok-4.3	1M largest	USD 1.25	USD 2.50	Lower-cost general model with a larger 1M context window
Grok 4.20 Reasoning grok-4.20-0309-reasoning	1M largest	USD 1.25	USD 2.50	Harder reasoning, document work and deliberate tool use
Grok 4.20 Non-reasoning grok-4.20-0309-non-reasoning	1M largest	USD 1.25	USD 2.50	High-volume text tasks that do not need explicit reasoning mode
Grok 4.20 Multi-agent grok-4.20-multi-agent-0309	1M largest	USD 1.25	USD 2.50	Hard multi-step work that benefits from parallel agent-style processing
Grok Build 0.1 grok-build-0.1	256k	USD 1.00 lowest input	USD 2.00	Coding-agent and build workflows at lower input cost

Prices in USD per 1,000,000 tokens. Last checked 2026-07-22. Re-check official figures on docs.x.ai.

How to choose

Grok 4.5 for broad quality

The current flagship and safer default for first builds. Grok 4.3 remains for the lowest price and the largest context window.

Grok Build 0.1 for coding agents

Test it when cost and repeated API calls matter for coding-agent workflows.

Test before you commit

Run the same realistic task against each candidate. Score correctness, latency, review effort and total token cost.

Keep a fallback plan

Route everyday requests to a lower-cost model and reserve larger context for hard cases. See the API limits guide.

Not building on the API? Use the plan chooser instead.

Open the calculator Models overview

Frequently asked questions

Estimate your own usage

Token volume matters more than model names once you move into production.

API calculator API guide

supergrok.info is an independent guide and resource site. It is not xAI, Grok, X, or an official login, billing, API, app, or support channel. For passwords, subscriptions, API keys, billing, app downloads, account access, incidents and support, use official xAI and Grok links. Grok and xAI are trademarks of their respective owners. This site uses those names only to describe and reference the product.

Grok models compared