Model Comparison

Grok-3 vs Grok-2

Grok-3 significantly outperforms across most benchmarks. Grok-2 is 1.5x cheaper per token.

Performance Benchmarks

Comparative analysis across standard metrics

2 benchmarks

Grok-3 outperforms in 2 benchmarks (GPQA, MMMU), while Grok-2 is better at 0 benchmarks.

Grok-3 significantly outperforms across most benchmarks.

Thu May 14 2026 • llm-stats.com

Arena Performance

Human preference votes

Pricing Analysis

Price comparison per million tokens

Grok-2 costs less

For input processing, Grok-3 ($3.00/1M tokens) is 1.5x more expensive than Grok-2 ($2.00/1M tokens).

For output processing, Grok-3 ($15.00/1M tokens) is 1.5x more expensive than Grok-2 ($10.00/1M tokens).

In conclusion, Grok-3 is more expensive than Grok-2.*

* Using a 3:1 ratio of input to output tokens

Lowest available price from all providers
Thu May 14 2026 • llm-stats.com
xAI
Grok-3
Input tokens$3.00
Output tokens$15.00
Best providerxAI
xAI
Grok-2
Input tokens$2.00
Output tokens$10.00
Best providerxAI
Notice missing or incorrect data?Start an Issue

Context Window

Maximum input and output token capacity

Both models have the same input context window of 128,000 tokens. Both models can generate responses up to 8,000 tokens.

xAI
Grok-3
Input128,000 tokens
Output8,000 tokens
xAI
Grok-2
Input128,000 tokens
Output8,000 tokens
Thu May 14 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Both Grok-3 and Grok-2 support multimodal inputs.

They are both capable of processing various types of data, offering versatility in application.

Grok-3

Text
Images
Audio
Video

Grok-2

Text
Images
Audio
Video

License

Usage and distribution terms

Both models are licensed under proprietary licenses.

Both models have usage restrictions defined by their respective organizations.

Grok-3

Proprietary

Closed source

Grok-2

Proprietary

Closed source

Release Timeline

When each model was launched

Grok-3 was released on 2025-02-17, while Grok-2 was released on 2024-08-13.

Grok-3 is 6 months newer than Grok-2.

Grok-3

Feb 17, 2025

1.2 years ago

6mo newer
Grok-2

Aug 13, 2024

1.8 years ago

Knowledge Cutoff

When training data ends

Grok-3 has a documented knowledge cutoff of 2024-11-17, while Grok-2's cutoff date is not specified.

We can confirm Grok-3's training data extends to 2024-11-17, but cannot make a direct comparison without Grok-2's cutoff date.

Grok-3

Nov 2024

Grok-2

Provider Availability

Grok-3 is available from xAI. Grok-2 is available from xAI.

Grok-3

xai logo
xAI
Input Price:Input: $3.00/1MOutput Price:Output: $15.00/1M

Grok-2

xai logo
xAI
Input Price:Input: $2.00/1MOutput Price:Output: $10.00/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion

Key Takeaways

Higher GPQA score (84.6% vs 56.0%)
Higher MMMU score (78.0% vs 66.1%)
Less expensive input tokens
Less expensive output tokens

Detailed Comparison

AI Model Comparison Table
Feature
xAI
Grok-3
xAI
Grok-2

FAQ

Common questions about Grok-3 vs Grok-2.

Which is better, Grok-3 or Grok-2?

Grok-3 significantly outperforms across most benchmarks. Grok-3 is made by xAI and Grok-2 is made by xAI. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does Grok-3 compare to Grok-2 in benchmarks?

Grok-3 scores AIME 2024: 93.3%, AIME 2025: 93.3%, GPQA: 84.6%, LiveCodeBench: 79.4%, MMMU: 78.0%. Grok-2 scores DocVQA: 93.6%, HumanEval: 88.4%, MMLU: 87.5%, MATH: 76.1%, MMLU-Pro: 75.5%.

Is Grok-3 cheaper than Grok-2?

Grok-2 is 1.5x cheaper for input tokens. Grok-3 costs $3.00/M input and $15.00/M output via xai. Grok-2 costs $2.00/M input and $10.00/M output via xai.

What are the context window sizes for Grok-3 and Grok-2?

Grok-3 supports 128K tokens and Grok-2 supports 128K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Grok-3 and Grok-2?

Key differences include input pricing ($3.00 vs $2.00/M). See the full comparison above for benchmark-by-benchmark results.