xAI Grok 4.3 API Pricing 2026: $1.25/$2.50 with 1M Context

Grok 4.3 is xAI's best-value general-purpose model — 1M context at $1.25/$2.50 per million tokens. Ideal for long-document analysis, codebase review, and cost-sensitive agentic workloads.

Input Price
$1.25/M
Under 200K tokens
Output Price
$2.50/M
Under 200K tokens
Context Window
1M tokens
Largest available from xAI
Best For
Long-context
Document analysis, RAG, code review

Published: Aug 22, 2026 | 95 models tracked across 11 providers

⚠️ Tiered Pricing: Grok 4.3 uses prompt-size tiered pricing. Under 200K tokens: $1.25/$2.50 per 1M tokens. 200K+ tokens: $2.50/$5.00 — double the base rate. Cached input: $0.20/M (under 200K) or $0.40/M (200K+).

Grok 4.3 Pricing Breakdown

Pricing at different monthly volumes (under 200K token prompts):

Monthly Volume Input Cost Output Cost Total (50/50)
10M tokens$6.25$12.50$18.75
50M tokens$31.25$62.50$93.75
100M tokens$62.50$125.00$187.50
500M tokens$312.50$625.00$937.50

At 100M tokens/month with a 50/50 input/output split, Grok 4.3 costs $187.50. Prompts over 200K tokens double the per-token cost.

xAI Grok Model Family Comparison

Model Input Output Context Best For
Grok 4.6$2.00/M$6.00/M500KFlagship tasks, complex reasoning
Grok 4.5$2.00/M$6.00/M500KGeneral-purpose, balanced
Grok 4.3$1.25/M$2.50/M1MLong-context, budget-friendly
Grok Build 0.1$1.00/M$2.00/M256KCode generation, API development

Grok 4.3 is xAI's cheapest model with a 1M context window. Grok Build 0.1 is slightly cheaper but limited to 256K context and optimized for code.

Grok 4.3 vs Competitors

Model Input Output Context Provider
Grok 4.3$1.25/M$2.50/M1MxAI
DeepSeek V4 Pro$0.66/M$1.98/M1MDeepSeek
GPT-5.4$2.50/M$15.00/M400KOpenAI
Claude Sonnet 5$2.00/M$10.00/M1MAnthropic
Gemini 3.1 Pro$2.00/M$12.00/M1MGoogle
Mistral Medium 3.5$1.50/M$7.50/M128KMistral
DeepSeek V4 Flash$0.22/M$0.66/M1MDeepSeek

Grok 4.3 is significantly cheaper than GPT-5.4 and Claude Sonnet 5, with a larger context window than GPT-5.4. DeepSeek V4 Pro is cheaper but Grok 4.3 offers xAI's reasoning capabilities.

Best Use Cases for Grok 4.3

1. Long-Document Analysis

With 1M context, Grok 4.3 can process entire books, legal contracts, or research papers in a single API call. At $1.25/M input, analyzing a 200-page document (~100K tokens) costs about $0.13.

2. Codebase Review and Refactoring

Grok 4.3's large context window lets you feed an entire codebase for review, bug detection, or refactoring suggestions. Cheaper than GPT-5.4 ($2.50/$15.00) for similar tasks.

3. RAG (Retrieval-Augmented Generation)

RAG pipelines benefit from large context windows to include more retrieved documents. Grok 4.3 at $1.25/M input makes it cost-effective for context-heavy RAG applications.

4. Multi-Turn Agentic Conversations

Agentic workflows that accumulate long conversation histories benefit from Grok 4.3's 1M context and low output price ($2.50/M). Much cheaper than Sonnet 5 ($10/M output) for extended agent sessions.

5. Data Extraction and Classification

Input-heavy tasks like extracting structured data from documents or classifying text are cost-effective at $1.25/M input. For even cheaper extraction, consider DeepSeek V4 Flash ($0.22/M).

6. Summarization at Scale

Summarizing large volumes of text is output-heavy but affordable at $2.50/M output. For a batch of 10,000 documents averaging 5K tokens each, the total cost is approximately $1.88 (input) + $12.50 (output at 500 tokens each) = $14.38.

Real-World Cost Scenario

Use case: 10,000 document summaries per month (5K input tokens, 500 output tokens each).

Component Tokens Rate Cost
Input tokens50M$1.25/M$62.50
Output tokens5M$2.50/M$12.50
Total55M—$75.00

The same workload on GPT-5.4 would cost $137.50 (83% more). On Claude Sonnet 5, it would cost $150 (100% more).

Frequently Asked Questions

How much does Grok 4.3 cost?
Grok 4.3 costs $1.25 per 1M input tokens and $2.50 per 1M output tokens for prompts under 200K tokens. For prompts at or above 200K tokens, the price doubles to $2.50/$5.00. Cached input tokens cost $0.20/M (under 200K) or $0.40/M (200K+).
What is the context window of Grok 4.3?
Grok 4.3 has a 1M token context window — the largest available from xAI and tied with Google, Anthropic, and DeepSeek for the largest context windows in the industry. This lets you process very long documents, codebases, or conversation histories in a single API call.
How does Grok 4.3 compare to GPT-5.4?
Grok 4.3 ($1.25/$2.50) is 50% cheaper on input and 83% cheaper on output than GPT-5.4 ($2.50/$15.00). Both have large context windows (1M for Grok 4.3, 400K for GPT-5.4). GPT-5.4 is a more capable mid-tier model, but Grok 4.3 offers dramatically better value for cost-sensitive workloads.
Is Grok 4.3 good for long-context tasks?
Yes. Grok 4.3's 1M token context window makes it excellent for long-context tasks like document analysis, codebase review, and multi-turn conversations. However, note that prompts over 200K tokens trigger tiered pricing ($2.50/$5.00 — double the base rate). For pure long-context budget work, DeepSeek V4 Flash ($0.22/$0.66, 1M context) is cheaper but less capable.
What is Grok 4.3's tiered pricing?
Grok 4.3 uses prompt-size tiered pricing. Prompts under 200K tokens cost $1.25/$2.50 per 1M tokens. Prompts at or above 200K tokens cost $2.50/$5.00 — exactly double. This means long-context workloads cost more per token. Plan your prompts to stay under 200K when possible to get the best rate.
Should I use Grok 4.3 or Grok Build 0.1?
Grok Build 0.1 ($1.00/$2.00, 256K context) is slightly cheaper and optimized for code generation. Grok 4.3 ($1.25/$2.50, 1M context) has 4x the context window and is a more general-purpose model. Choose Grok Build 0.1 for budget code tasks with short contexts. Choose Grok 4.3 when you need long context or broader capabilities.

Compare 95 Models Across 11 Providers

See how Grok 4.3 stacks up against every major AI model on price, context window, and capabilities.

View All Model Pricing

Related Pages