GPT-5 High vs Grok-3 Mini: Specs & Benchmark Comparison

GPT-5 High is developed by OpenAI, while Grok-3 Mini comes from xAI. Grok-3 Mini was released in February 2025, and GPT-5 High followed 6 months later in August 2025.

The two models share 2 published benchmarks. GPT-5 High leads on 2 of them. The widest gaps are on AIME 2025, where GPT-5 High scores 95.0% against 91.0%; GPQA, where GPT-5 High scores 87.0% against 84.0%. Averaged across everything we track, GPT-5 High sits at 91.0% and Grok-3 Mini at 87.8%.

Grok-3 Mini is the cheaper API at $0.30 per million input tokens and $0.50 per million output tokens, roughly 4 times cheaper than GPT-5 High at $1.25 and $10. GPT-5 High takes the larger context window at 400K tokens, compared with 128K for Grok-3 Mini. Both accept text and images as input.

CharacteristicGPT-5 HighGrok-3 Mini
CompanyOpenAIxAI
Release DateAugust 6, 2025February 16, 2025
Parameters——
MultimodalYesYes
Context (input)400K128K
Context (output)128K8K
Input Price / 1M$1.25$0.30
Output Price / 1M$10.00$0.50
Average Score91.0%87.8%
Benchmarks
AIME 202595.0%91.0%
GPQA87.0%84.0%

Visual Benchmark Comparison

GPT-5 High
Grok-3 Mini
AIME 20250.9 vs 0.9
0.9
0.9
GPQA0.9 vs 0.8
0.9
0.8

Verdict

GPT-5 High leads in 2 out of 4 comparison categories.

Overall Performance

Both models show comparable average scores: GPT-5 High — 0.9, Grok-3 Mini — 0.9.

API Cost

Grok-3 Mini is 14.1x cheaper: input $0.30/1M vs $1.25/1M tokens.

Context Window

GPT-5 High supports a larger context: 400K vs 128K tokens.

Recency

GPT-5 High is newer: released 8/6/2025 vs 2/16/2025.

More About These Models

Related Comparisons

Frequently Asked Questions

Which is better for coding — GPT-5 High or Grok-3 Mini?
Direct comparison on the SWE-Bench benchmark is not available. We recommend reviewing other metrics on the comparison page.
Which model is cheaper — GPT-5 High or Grok-3 Mini?
Grok-3 Mini is cheaper for input: $0.30 per 1M tokens vs $1.25.
Which has a larger context window — GPT-5 High or Grok-3 Mini?
GPT-5 High supports a larger context: 400,000 tokens vs 128,000.

The GPT-5 High and Grok-3 Mini comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the GPT-5 High or Grok-3 Mini page. See also the complete list of AI model comparisons.