// gpt
GPT-5
OpenAI's flagship. Sharper reasoning, higher price.
Not currently callable on llmrelay
Our upstream does not serve gpt-5, so calls fail rather than
bill. The rates below are the official list price and what our 50% tier would charge, kept here for
reference.
For work you can run today, GPT-5.6 Sol is live at
$2.5/M input.
Official list
llmrelay
Input / M tokens
$5.00
$2.50
Output / M tokens
$20.00
$10.00
Context window
—
400,000 tokens
Max output
—
16,384 tokens
What it costs you per month
Real-world budget scenarios. Numbers are simple sums — official list price vs llmrelay's 50% off tier.
Usage scenario
Official
llmrelay
Light coding (1M in / 200K out per month)
$9.00
$4.50save $4.50
Heavy Cursor / Cline user (50M in / 5M out per month)
$350.00
$175.00save $175.00
Production RAG (500M in / 20M out per month)
$2900.00
$1450.00save $1450.00
What it's good at
- +Reasoning depth
- +Function calling ecosystem
- +Wide tool support
Best for
- — OpenAI-native agents
- — Complex reasoning benchmarks
- — Structured planning
Run GPT-5.6 Sol instead
GPT-5 is not callable on llmrelay today, so we are not going to sell you an API key for it. GPT-5.6 Sol is live at $2.5/M input, also 50% off list.
View GPT-5.6 Sol →