Skip to content
Sovyron

GPT-5.2

gpt400k context128k max outputreleased Dec 11, 2025ReasoningTool calling

As of Nov 22, 2026, the cheapest GPT-5.2 API price is $1.5909 per 1M input tokens (Poe) and $12.7273 per 1M output tokens (Poe). GPT-5.2 is offered by 33 providers in the gpt family, reasoning enabled. OpenAI's own rate is $14 per 1M output tokens.

Cheapest input
$1.5909
Cheapest output
$12.7273
Official (OpenAI)
$14
output per 1M tokens
Providers
33
offering this model
Spread
13.2×
max / min output
Free offers
0
free-tier offers

API pricing by provider

Metered per-token prices. Cheapest input and output are highlighted; † marks a context-tier surcharge. How prices are ranked. Seeing a wide range? Why the same LLM costs different amounts.

GPT-5.2 API prices by provider in USD per 1M tokens, updated Nov 22, 2026
Poeopenai/gpt-5.2Cheapest$1.5909Cheapest$12.7273$0.1591—400k31d ago
Jiekou.AIgpt-5.2$1.6625$13.3$0.1662—400k31d ago
302.AIgpt-5.2$1.75$14——400k—
Abacusgpt-5.2$1.75$14$0.175—400k—
AIHubMixgpt-5.2$1.75$14$0.175—400k—
AI/ML APIopenai/gpt-5-2$1.75$14$0.175—400k30d agoout −23.1%
Azuregpt-5.2$1.75$14$0.175—400k—
Azure Cognitive Servicesgpt-5.2$1.75$14$0.175—400k—
Databricksdatabricks-gpt-5-2$1.75$14$0.175—400k—
DigitalOceanopenai-gpt-5.2$1.75$14$0.175—400k—
Eden AIopenai/gpt-5.2$1.75$14$0.175—400k—
FastRouteropenai/gpt-5.2$1.75$14$0.175—400k—
Impossiblopenai/gpt-5.2$1.75$14$0.175—400k—
Kilo Gatewayopenai/gpt-5.2$1.75$14$0.175—400k—
DevPass (LLM Gateway)gpt-5.2$1.75$14$0.175—400k—
LLM Gatewayazure/gpt-5.2$1.75$14$0.175—400k—
LLM Gatewayopenai/gpt-5.2$1.75$14$0.175—400k—
LowRouterauto/openai/gpt-5.2$1.75$14$0.175—400k—
Merge Gatewayopenai/gpt-5.2$1.75$14$0.175—400k—
NanoGPTopenai/gpt-5.2$1.75$14$0.175—400k—
NEAR AI Cloudopenai/gpt-5.2$1.75$14$0.175—400k—
Neongpt-5-2$1.75$14$0.175—400k—
Ofoxopenai/gpt-5.2$1.75$14$0.18—400k36d ago
OpenAIgpt-5.2$1.75$14$0.175—400k—
OpenCodegpt-5.2$1.75$14$0.175—400k—
OpenRouteropenai/gpt-5.2$1.75$14$0.175—400k—
OrcaRouteropenai/gpt-5.2$1.75$14$0.175—400k—
Prime Inferenceopenai/gpt-5.2$1.75$14——400k—
SAP AI Coregpt-5.2$1.75$14$0.175—400k31d ago
Snowflake Cortexopenai-gpt-5.2$1.75$14$0.18—400k—
Venice AIopenai-gpt-52$1.75$14$0.175—256k37d ago
Vercel AI Gatewayopenai/gpt-5.2$1.75$14$0.175—400k—
ZenMuxopenai/gpt-5.2$1.75$14$0.175—400k—
Vercel AI Gatewayopenai/gpt-5.2-pro$21$168——400k—

GPT-5.2 price history

Output price per 1M tokens since tracking began on Sep 10, 2026, for the 6 providers whose rate moved. Each step is a recorded change. All recent price changes.

GPT-5.2 output price per 1M tokens by provider, Sep 10, 2026 to Nov 22, 2026$0$5$10$15$20Sep 10, 2026Oct 17, 2026Nov 22, 2026
Poe$12.7273Jiekou.AI$13.3AI/ML API$14Ofox$14SAP AI Core$14Venice AI$14
GPT-5.2 price changes by provider, newest first, USD per 1M tokens
DateProvider1M input1M outputChange
AI/ML API$2.275 → $1.75$18.2 → $14−23.1%
SAP AI Core$1.25 → $1.75$9.44 → $14+48.3%
Jiekou.AI$1.575 → $1.6625$12.6 → $13.3+5.6%
Poe$1.6 → $1.5909$13 → $12.7273−2.1%
Ofox$1.4 → $1.75$11.2 → $14+25%
Venice AI$2.19 → $1.75$17.5 → $14−20%
Ofox$1.75 → $1.4$14 → $11.2−20%

GPT-5.2 pricing FAQ

What is the cheapest GPT-5.2 API?

As of Nov 22, 2026, Poe has the lowest GPT-5.2 output price at $12.7273 per 1M tokens, and Poe has the lowest input price at $1.5909 per 1M tokens. Prices are metered per-token rates in USD; free tiers and subscription plans are excluded.

How much does GPT-5.2 cost on OpenAI?

OpenAI charges $1.75 per 1M input tokens and $14 per 1M output tokens for GPT-5.2.

How many providers offer GPT-5.2?

33 providers list GPT-5.2 on Sovyron; 34 of them sell it at a metered per-token price.

Is GPT-5.2 free?

No provider in the Sovyron catalog lists a free tier for GPT-5.2.

What is the context window of GPT-5.2?

GPT-5.2 supports a 400k-token context window and up to 128k output tokens.

Token cost calculator

What a month costs at one provider's published rates. Enter the tokens you expect to send and receive; cached tokens use the cached-input price where the provider lists one, and count as ordinary input where it does not. See how much prompt caching saves and why output tokens cost more than input tokens.

Estimated

—