GPT-5.1 vs GPT-5.4

Side-by-side benchmark comparison across coding, math, reasoning, speed, and pricing.

GPT-5.4 by OpenAI wins on 5 of 9 benchmarks against GPT-5.1 by OpenAI, which leads on 3. This head-to-head comparison covers coding, math, reasoning, speed, and pricing metrics from our benchmark data.

Category-by-Category Breakdown

Coding: In coding, GPT-5.4 scores 71.1 on AA Coding Index compared to GPT-5.1's 49.4, while GPT-5.4 scores 1201 on Design Arena ELO compared to GPT-5.1's 1158, while GPT-5.4 scores 1210 on Website Arena ELO compared to GPT-5.1's 1179.

Reasoning: In reasoning, GPT-5.4 scores 90.3% on GPQA Diamond compared to GPT-5.1's 86.2%, while both score 75.7% on τ²-Bench Airline.

Context: In context, GPT-5.4 scores 1.1M on Context Length compared to GPT-5.1's 400K.

Pricing Comparison

GPT-5.1 costs $1.3/1M input tokens and $10.0/1M output tokens, while GPT-5.4 costs $2.5/1M input and $15.0/1M output. GPT-5.1 is the more affordable option for API usage.

Verdict

GPT-5.1 leads across the board in affordability, making it the stronger overall choice in this comparison.

View Individual Model Pages

GPT-5.1 vs GPT-5.4 — FAQ

Which is better, GPT-5.1 or GPT-5.4?

GPT-5.4 wins on more benchmarks overall (5 vs 3). However, the best choice depends on your specific needs — each model excels in different areas.

How does GPT-5.1 compare to GPT-5.4 for coding?

SWE-bench Verified data is not available for both models. Check the detailed comparison charts above for other coding-related metrics.

Is GPT-5.1 cheaper than GPT-5.4?

Yes, GPT-5.1 is cheaper. GPT-5.1 costs $1.3/1M input and $10.0/1M output tokens. GPT-5.4 costs $2.5/1M input and $15.0/1M output tokens.

Which is faster, GPT-5.1 or GPT-5.4?

Output speed data is not available for both models. Check the speed section of the comparison above for available performance data.

What benchmarks does the GPT-5.1 vs GPT-5.4 comparison cover?

This comparison covers 9 benchmarks including GPQA Diamond, Input Cost, Output Cost, Context Length, AA Coding Index, Cached Input Cost, Design Arena ELO, τ²-Bench Airline, and more. Metrics span general intelligence, coding, math, reasoning, speed, and cost categories.