Requesty

gpt-5 vs claude-sonnet-4-6

Side-by-side comparison of gpt-5 and claude-sonnet-4-6— benchmarks, pricing, context window and capabilities. Both are accessible through Requesty's unified API. claude-sonnet-4-6 outperforms gpt-5 on 6 of 7 shared benchmarks.

Benchmark comparison

Intelligence Indexreasoning
gpt-544.6%
claude-sonnet-4-651.7%
Coding Indexcoding
gpt-536.0%
claude-sonnet-4-650.9%
Math Indexmath
gpt-594.3%
claude-sonnet-4-6
GPQA Diamondreasoning
gpt-585.4%
claude-sonnet-4-687.5%
AIME 2025math
gpt-594.3%
claude-sonnet-4-6
LiveCodeBenchcoding
gpt-584.6%
claude-sonnet-4-6
Terminal-Bench Hardagentic
gpt-532.6%
claude-sonnet-4-653.0%
τ²-Benchagentic
gpt-584.8%
claude-sonnet-4-675.7%
SciCodecoding
gpt-542.9%
claude-sonnet-4-646.8%
MMLU Proknowledge
gpt-587.1%
claude-sonnet-4-6
Humanity's Last Examreasoning
gpt-526.5%
claude-sonnet-4-630.0%

Scores sourced from official model cards, Artificial Analysis, and public leaderboards. Benchmarks measure specific skills and don't capture every aspect of model quality.

Pricing & specifications

gpt-5claude-sonnet-4-6
Input price / 1M$1.25$3.00
Output price / 1M$10.00$15.00
Context window400K tokens1M tokens
Max output128K tokens64K tokens
Vision inputYesYes
Tool callingYesYes
ReasoningYesYes
Prompt cachingYesYes
Computer useYes
ProviderOpenAI Inc.Anthropic PBC

Questions people ask

Is gpt-5 better than claude-sonnet-4-6?
claude-sonnet-4-6 outperforms gpt-5 on 6 of 7 shared benchmarks. See the benchmark comparison above for specifics — gpt-5 and claude-sonnet-4-6 have different strengths across reasoning, coding, math and multimodal tasks.
Which is cheaper — gpt-5 or claude-sonnet-4-6?
gpt-5 is cheaper. gpt-5 costs $1.25/$10.00 per 1M input/output tokens, while claude-sonnet-4-6 costs $3.00/$15.00.
Can I use gpt-5 and claude-sonnet-4-6 through the same API?
Yes. Requesty provides a single OpenAI-compatible API that routes to both. Change just the "model" parameter to switch — "openai/gpt-5" or "anthropic/claude-sonnet-4-6" — no other code changes needed.
What are the context windows?
gpt-5 supports up to 400K tokens of context. claude-sonnet-4-6 supports up to 1M tokens. Longer context means you can feed larger documents or codebases in a single prompt, though quality often degrades past 128K for most models.

Switch between gpt-5 and claude-sonnet-4-6 with one line of code

Requesty provides a single OpenAI-compatible API for 400+ models. Change the model parameter, not your code.

Get started free