LLM API pricing / xAI
What the Grok API costs, model by model
Short answer. xAI lists 8 models with a per-token price on its API. Input: from $1.00 (Grok Build 0.1) to $2.00 (Grok 4.5, Grok 4.6 and Grok 4.7) per 1M tokens. Output: from $2.00 (Grok Build 0.1) to $6.00 (Grok 4.5, Grok 4.6 and Grok 4.7). Read on xAI's own pages on October 26, 2026.
In plain words. You pay for the tokens you send (input) and the tokens the model writes (output). One request of 10,000 tokens in and 1,000 out costs $0.012 on Grok Build 0.1 and $0.026 on Grok 4.5, Grok 4.6 and Grok 4.7. Output costs more than input on all 8 models.
Paying for the app? See Grok plans and prices →
Current models
8 models, cheapest input first, verified on October 26, 2026. The last column is the cost of one request of 10,000 input and 1,000 output tokens at these prices, with no cache.
| Model | Input $/1M | Output $/1M | Cached input | Context | Example request | Verified |
|---|---|---|---|---|---|---|
| Grok Build 0.1 | $1.00 | $2.00 | $0.20 | 256K | $0.012 | 2026-10-26 |
| Grok 4.20 (non-reasoning) | $1.25 | $2.50 | $0.20 | 1M | $0.015 | 2026-10-26 |
| Grok 4.20 (reasoning) | $1.25 | $2.50 | $0.20 | 1M | $0.015 | 2026-10-26 |
| Grok 4.20 Multi-agent | $1.25 | $2.50 | $0.20 | 1M | $0.015 | 2026-10-26 |
| Grok 4.3 | $1.25 | $2.50 | $0.20 | 1M | $0.015 | 2026-10-26 |
| Grok 4.5 | $2.00 | $6.00 | $0.30 | 500K | $0.026 | 2026-10-26 |
| Grok 4.6 | $2.00 | $6.00 | $0.50 | 500K | $0.026 | 2026-10-26 |
| Grok 4.7 | $2.00 | $6.00 | $0.50 | 500K | $0.026 | 2026-10-26 |
How xAI describes its models
- Grok Build 0.1:
SpaceXSI's intelligent coding model for agentic software, engineering, and workflow tasks.
(docs.x.ai, read October 19, 2026) - Grok 4.20 (non-reasoning):
Grok 4.20 is a high-performance model with industry-leading speed and agentic tool calling capabilities.
(docs.x.ai, read September 30, 2026) - Grok 4.20 (reasoning):
Grok 4.20 is a high-performance model with industry-leading speed and agentic tool calling capabilities.
(docs.x.ai, read September 30, 2026) - Grok 4.20 Multi-agent:
Multiple agents collaborate in parallel to perform deep research tasks.
(docs.x.ai, read September 30, 2026) - Grok 4.3:
Fast, reliable model with strong tool calling and instruction following capabilities.
(docs.x.ai, read September 30, 2026) - Grok 4.5:
SpaceXSI's intelligent coding model for agentic software, engineering, and workflow tasks.
(docs.x.ai, read October 19, 2026) - Grok 4.6:
SpaceXSI's frontier model for coding, agentic tasks, and knowledge work.
(docs.x.ai, read October 19, 2026) - Grok 4.7:
SpaceXSI's frontier model for coding, agentic tasks, and knowledge work.
(docs.x.ai, read October 19, 2026)
How xAI prices its models
- 8 of the 8 models publish a cached-input price, at 15%, 16%, 20% or 25% of the input price.
- 8 models charge more for long prompts (prompts of 200,000 tokens or more): Grok Build 0.1 $2.00 / $4.00, Grok 4.20 (non-reasoning) $2.50 / $5.00, Grok 4.20 (reasoning) $2.50 / $5.00, Grok 4.20 Multi-agent $2.50 / $5.00, Grok 4.3 $2.50 / $5.00, Grok 4.5 $4.00 / $12.00, Grok 4.6 $4.00 / $12.00, Grok 4.7 $4.00 / $12.00.
Price lists with no figure per model
Batch
Condition: wait: completion is best effort, most within 24 hours. No figure is shown here: the per-model batch prices sit behind a toggle on each model's page that this site could not read without a browser, so no figure is shown.
The Batch API lets you process large volumes of requests asynchronously at a discount to standard pricing. The size of the discount varies by model.
(docs.x.ai, read October 26, 2026)The batch discount applies to all token types — input tokens, output tokens, cached tokens, and reasoning tokens.
(docs.x.ai, read October 26, 2026)Batch discounts by model: Model | Batch discount | grok-4.7 grok-4.6 grok-4.5 grok-4.3 grok-4.20-0309-reasoning grok-4.20-0309-non-reasoning grok-4.20-multi-agent-0309 | 20 % |
(docs.x.ai, read October 26, 2026)
What xAI's own pages say
Batch API
- Grok 4.7 and Grok 4.6:
Batch API Supported in us-east-1, us-west-2
(docs.x.ai, docs.x.ai, read October 26, 2026) - Grok 4.5, Grok 4.3, Grok 4.20 (reasoning), Grok 4.20 (non-reasoning) and Grok 4.20 Multi-agent:
Batch API Supported
(docs.x.ai, docs.x.ai, docs.x.ai, docs.x.ai, docs.x.ai, read October 26, 2026) - Grok Build 0.1:
Batch API Not supported
(docs.x.ai, read October 26, 2026)
Long prompts
- xAI's model page for Grok 4.3:
We charge different rates for requests which exceed the 200K context window
(docs.x.ai, read October 26, 2026)
Retirements
Retired: Grok 3 (May 15, 2026), Grok 4 (May 15, 2026), Grok 4 Fast (non-reasoning) (May 15, 2026), Grok 4 Fast (reasoning) (May 15, 2026), Grok 4.1 Fast (non-reasoning) (May 15, 2026), Grok 4.1 Fast (reasoning) (May 15, 2026).
- May 15, 2026: Retirement of the whole grok-4 and grok-3 families source
Analysis that cites xAI models
- Prompt Caching Is Not 10% Everywhere: What a Cache Hit Costs Across Eight Providers · names 4 xAI models
- Same List Price, Different Bill: Five Pairs of LLM APIs Where the Workload Decides What You Pay · names 1 xAI model
- The Second Price List: What Batch, Flex, Priority and Data-Sharing Discounts Cost Across Seven Providers · names 1 xAI model
- Five Things Decide What You Pay for a Model, and Only One of Them Is the Model · names 1 xAI model
- Grok vs ChatGPT: Price in Spain, Plan by Plan · October 20, 2026
- Cursor Pricing Includes Grok and X: Plans From $8 to $200 · October 19, 2026
FAQ
What is the cheapest Grok model on the API?
Grok Build 0.1, at $1.00 per 1M input tokens and $2.00 per 1M output tokens, as xAI's pricing page printed it on October 26, 2026.
Which Grok models work with the xAI Batch API?
Grok 4.7 and Grok 4.6: “Batch API Supported in us-east-1, us-west-2”. Grok 4.5, Grok 4.3, Grok 4.20 (reasoning), Grok 4.20 (non-reasoning) and Grok 4.20 Multi-agent: “Batch API Supported”. Grok Build 0.1: “Batch API Not supported”. Read on October 26, 2026.
Do Grok prices change for long prompts?
xAI's model page for Grok 4.3: “We charge different rates for requests which exceed the 200K context window”. Read on October 26, 2026. On xAI's pricing page, all 8 models have a second price for prompts of 200,000 tokens or more.
How we checked
- Each price was read on xAI's pricing page (docs.x.ai) or on a model page it links. The Verified column links to that page and gives the day.
- Every Monday a script reads xAI's pricing page again and compares each row with the figures here. A difference is read by hand before a price changes.
- Quotes are copied as xAI's pages print them, with the day they were read. A script checks every week that the model descriptions and the quotes under “What xAI's own pages say” are still on their pages.
- The example request is worked out from the per-1M prices: 10,000 input tokens and 1,000 output tokens, no cache, no batch discount, no long-prompt tier.
Full method: methodology. All providers: the full pricing table.