OpenAI

GPT API

OpenAI's GPT line — GPT-5.x and GPT-6 tiers priced per million tokens, with a spread of more than sixty times.

models in the line

At a glance

From: $0.12 / 1M tokens

Models in this family

Model	From	Modes	Open
GPT 6 LunaFast, low-cost model of the GPT-6 series for chat, classification and high-volume tasks. Text, image and file input.	$0.12/ 1M tokens	—	Open Docs
GPT 5.6 LunaFast, cost-efficient model of the GPT-5.6 series. Suited for high-volume, latency-sensitive chat, classification and lightweight agentic workflows.	$0.18/ 1M tokens	—	Open Docs
GPT 5.6 TerraHigh-capability GPT-5.6 model for long-context, tool-enabled workflows across writing, coding, analysis and multi-step problem solving. Text, image and file input.	$1.2/ 1M tokens	—	Open Docs
GPT 5.5Broadly capable GPT-5.5 model for everyday reasoning, coding and writing.	$1.5/ 1M tokens	—	Open Docs
GPT 6 SolHigh-capability model of the GPT-6 series for demanding reasoning, coding and agentic tasks. Text, image and file input.	$2.4/ 1M tokens	—	Open Docs
GPT 6.1 SolRefreshed Sol model of the GPT-6.1 series for demanding reasoning, coding and agentic tasks. Text, image and file input.	$1.8/ 1M tokens	—	Open Docs
GPT 5.6 SolFrontier model of the GPT-5.6 series for the hardest reasoning, coding and agentic tasks. Text, image and file input, 1M-token context.	$3/ 1M tokens	—	Open Docs
GPT 6 AstraFrontier GPT-6 model for the hardest reasoning, coding and agentic work. Text, image and file input.	$8/ 1M tokens	—	Open Docs

Pricing

Model	Option	Price
GPT 6 Luna	input	$0.12/ 1M tokens
GPT 6 Luna	output	$0.6/ 1M tokens
GPT 6 Luna	cacheRead	$0.012/ 1M tokens
GPT 6 Luna	cacheWrite	$0.15/ 1M tokens
GPT 5.6 Luna	input	$0.18/ 1M tokens
GPT 5.6 Luna	output	$1.08/ 1M tokens
GPT 5.6 Luna	cacheRead	$0.018/ 1M tokens
GPT 5.6 Luna	cacheWrite	$0.225/ 1M tokens
GPT 5.6 Terra	input	$1.2/ 1M tokens
GPT 5.6 Terra	output	$7.2/ 1M tokens
GPT 5.6 Terra	cacheRead	$0.12/ 1M tokens
GPT 5.6 Terra	cacheWrite	$1.5/ 1M tokens
GPT 5.5	input	$1.5/ 1M tokens
GPT 5.5	output	$9/ 1M tokens
GPT 5.5	cacheRead	$0.15/ 1M tokens
GPT 6 Sol	input	$2.4/ 1M tokens
GPT 6 Sol	output	$12/ 1M tokens
GPT 6 Sol	cacheRead	$0.24/ 1M tokens
GPT 6 Sol	cacheWrite	$3/ 1M tokens
GPT 6.1 Sol	input	$1.8/ 1M tokens
GPT 6.1 Sol	output	$9/ 1M tokens
GPT 6.1 Sol	cacheRead	$0.18/ 1M tokens
GPT 6.1 Sol	cacheWrite	$2.25/ 1M tokens
GPT 5.6 Sol	input	$3/ 1M tokens
GPT 5.6 Sol	output	$18/ 1M tokens
GPT 5.6 Sol	cacheRead	$0.3/ 1M tokens
GPT 5.6 Sol	cacheWrite	$3.75/ 1M tokens
GPT 6 Astra	input	$8/ 1M tokens
GPT 6 Astra	output	$40/ 1M tokens
GPT 6 Astra	cacheRead	$0.8/ 1M tokens
GPT 6 Astra	cacheWrite	$10/ 1M tokens

Which one to pick

GPT 6 Luna
The cheapest tier in the line, about a third below 5.6 Luna on input — high-volume chat, classification and light agentic steps.
GPT 5.6 Luna
High-volume, latency-sensitive work — chat, classification, light agentic steps. A small fraction of Sol's input price.
GPT 5.6 Terra
Long-context, tool-enabled workflows across writing, coding and analysis. The middle of the line.
GPT 5.5
The previous generation, for everyday reasoning, coding and writing.
GPT 6 Sol
The GPT-6 generation of the high-capability tier, for demanding reasoning, coding and agentic tasks — cheaper than 5.6 Sol on every rate.
GPT 6.1 Sol
The refreshed 6.1 revision of Sol at the same price as GPT-6 Sol — the pick when you want the newest weights.
GPT 5.6 Sol
The hardest reasoning, coding and agentic tasks, with a 1M-token context and file input.
GPT 6 Astra
The frontier GPT-6 tier for the hardest reasoning, coding and agentic work — and the most expensive model in the line.

Strengths

Cached input is billed at one tenth of fresh input on every tier, which pays back immediately on a fixed system prompt.
One protocol across the whole line — moving between tiers or generations is a model-name change.
5.6 Sol carries a 1M-token context, so long documents do not need chunking.
Two generations side by side: GPT-6 Luna and Sol undercut their 5.6 counterparts, so at those tiers the newer model is also the cheaper one.

Limitations

Output tokens cost five to six times input on every tier, so verbose responses dominate the bill.
Cache writes are billed above fresh input on the 5.6 and GPT-6 tiers, and GPT-5.5 lists no cache-write rate — so caching only pays back on reuse.
Astra is priced at more than three times GPT-6 Sol, so it needs a task that justifies the premium.
Text-only in and out here — image and video generation come from other families.

Frequently asked questions

How much does the GPT API cost?

API Stock bills per use from one prepaid balance, with no subscription. Current prices: GPT 6 Luna — $0.12 per 1M input tokens and $0.6 per 1M output tokens; GPT 5.6 Luna — $0.18 per 1M input tokens and $1.08 per 1M output tokens; GPT 5.6 Terra — $1.2 per 1M input tokens and $7.2 per 1M output tokens; GPT 5.5 — $1.5 per 1M input tokens and $9 per 1M output tokens; GPT 6 Sol — $2.4 per 1M input tokens and $12 per 1M output tokens; GPT 6.1 Sol — $1.8 per 1M input tokens and $9 per 1M output tokens; GPT 5.6 Sol — $3 per 1M input tokens and $18 per 1M output tokens; GPT 6 Astra — $8 per 1M input tokens and $40 per 1M output tokens.

Which GPT model is the cheapest?

GPT 6 Luna is the cheapest at $0.12 per 1M input tokens. The most expensive tier, GPT 6 Astra, costs $8 per 1M input tokens.

Does the GPT API support prompt caching?

Yes. Cached input tokens are billed at a lower rate: GPT 6 Luna — $0.012 instead of $0.12 per 1M tokens; GPT 5.6 Luna — $0.018 instead of $0.18 per 1M tokens; GPT 5.6 Terra — $0.12 instead of $1.2 per 1M tokens; GPT 5.5 — $0.15 instead of $1.5 per 1M tokens; GPT 6 Sol — $0.24 instead of $2.4 per 1M tokens; GPT 6.1 Sol — $0.18 instead of $1.8 per 1M tokens; GPT 5.6 Sol — $0.3 instead of $3 per 1M tokens; GPT 6 Astra — $0.8 instead of $8 per 1M tokens.

Can I use the GPT API with the OpenAI or Anthropic SDK?

Yes. GPT models accept the OpenAI request format: set the SDK's base URL to https://api.api-stock.com/api/v1 and use your API Stock API key. Switching between models only changes the model id.

How do I get an API key for GPT?

Create an account and API key on API Stock, then fund your balance. The same key works with GPT and other catalog models.

The vendor's own description of this line: OpenAI — models