Compare/GPT-5.6 Terra (medium) vs Claude 3 Sonnet

GPT-5.6 Terra (medium)vsClaude 3 Sonnet

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

OpenAI

GPT-5.6 Terra (medium)

Input
$2.5/M
Output
$15/M
Speed
145 tok/s
TTFT
1.00s
Anthropic

Claude 3 Sonnet

Input
$3/M
Output
$15/M
Speed
TTFT

Winner by Category

Cheaper
GPT-5.6 Terra (medium)
Faster (tok/s)
GPT-5.6 Terra (medium)
Lower Latency
Claude 3 Sonnet
Benchmarks (6-3)
GPT-5.6 Terra (medium)

Pricing Comparison

MetricGPT-5.6 Terra (medium)Claude 3 Sonnet
Input ($/M tokens)$2.5$3
Output ($/M tokens)$15$15
Cost for 1M input + 100K output tokens:
GPT-5.6 Terra (medium)$4.00
Claude 3 Sonnet$4.50

Speed Comparison

Output Speed (tokens/s) — higher is better
GPT-5.6 Terra (medium)
145 tok/s
Claude 3 Sonnet
Time to First Token (seconds) — lower is better
GPT-5.6 Terra (medium)
1.00s
Claude 3 Sonnet

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
45.64.7
Coding Index
64.7
Math Index
GPQA Diamond
87.2%40.0%
MMLU-Pro
57.9%
LiveCodeBench
17.5%
AIME 2025
MATH-500
41.4%
Humanity's Last Exam
31.6%3.8%
SciCode
49.7%22.9%
IFBench
62.2%
TerminalBench
GPT-5.6 Terra (medium)6 wins
3 winsClaude 3 Sonnet

Frequently Asked Questions

Which is cheaper, GPT-5.6 Terra (medium) or Claude 3 Sonnet?

GPT-5.6 Terra (medium) is cheaper overall. Its blended price (3:1 input/output ratio) is $5.63/M tokens vs $6.00/M for Claude 3 Sonnet.

Which model performs better on benchmarks?

GPT-5.6 Terra (medium) wins 6 out of 12 benchmarks compared to 3 for Claude 3 Sonnet. See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

GPT-5.6 Terra (medium) generates tokens faster at 145 tok/s vs 0 tok/s. However, Claude 3 Sonnet has lower time-to-first-token (0.00s vs 1.00s).

When should I use GPT-5.6 Terra (medium) vs Claude 3 Sonnet?

Choose based on your priorities: GPT-5.6 Terra (medium) for lower cost, GPT-5.6 Terra (medium) for stronger benchmark performance, and GPT-5.6 Terra (medium) for faster generation. For latency-sensitive apps, check the TTFT comparison above.