Compare/Claude 4.5 Sonnet (Reasoning) vs GPT-5.4 (Non-reasoning)

Claude 4.5 Sonnet (Reasoning)vsGPT-5.4 (Non-reasoning)

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

Anthropic

Claude 4.5 Sonnet (Reasoning)

Input
$3/M
Output
$15/M
Speed
58 tok/s
TTFT
7.49s
OpenAI

GPT-5.4 (Non-reasoning)

Input
$2.5/M
Output
$15/M
Speed
61 tok/s
TTFT
0.57s

Winner by Category

Cheaper
GPT-5.4 (Non-reasoning)
Faster (tok/s)
GPT-5.4 (Non-reasoning)
Lower Latency
GPT-5.4 (Non-reasoning)
Benchmarks (8-3)
Claude 4.5 Sonnet (Reasoning)

Pricing Comparison

MetricClaude 4.5 Sonnet (Reasoning)GPT-5.4 (Non-reasoning)
Input ($/M tokens)$3$2.5
Output ($/M tokens)$15$15
Cost for 1M input + 100K output tokens:
Claude 4.5 Sonnet (Reasoning)$4.50
GPT-5.4 (Non-reasoning)$4.00

Speed Comparison

Output Speed (tokens/s) — higher is better
Claude 4.5 Sonnet (Reasoning)
58 tok/s
GPT-5.4 (Non-reasoning)
61 tok/s
Time to First Token (seconds) — lower is better
Claude 4.5 Sonnet (Reasoning)
7.49s
GPT-5.4 (Non-reasoning)
0.57s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
43.035.4
Coding Index
38.641.0
Math Index
88.0
GPQA Diamond
83.4%74.8%
MMLU-Pro
87.5%
LiveCodeBench
71.4%
AIME 2025
88.0%
MATH-500
Humanity's Last Exam
17.3%10.6%
SciCode
44.7%47.1%
IFBench
57.3%48.4%
TerminalBench
35.6%37.9%
Claude 4.5 Sonnet (Reasoning)8 wins
3 winsGPT-5.4 (Non-reasoning)

Frequently Asked Questions

Which is cheaper, Claude 4.5 Sonnet (Reasoning) or GPT-5.4 (Non-reasoning)?

GPT-5.4 (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $5.63/M tokens vs $6.00/M for Claude 4.5 Sonnet (Reasoning).

Which model performs better on benchmarks?

Claude 4.5 Sonnet (Reasoning) wins 8 out of 12 benchmarks compared to 3 for GPT-5.4 (Non-reasoning). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

GPT-5.4 (Non-reasoning) generates tokens faster at 61 tok/s vs 58 tok/s. However, GPT-5.4 (Non-reasoning) has lower time-to-first-token (0.57s vs 7.49s).

When should I use Claude 4.5 Sonnet (Reasoning) vs GPT-5.4 (Non-reasoning)?

Choose based on your priorities: GPT-5.4 (Non-reasoning) for lower cost, Claude 4.5 Sonnet (Reasoning) for stronger benchmark performance, and GPT-5.4 (Non-reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.