Does the discount apply if I call OpenAI's API directly?

No. The GPT-5.6 Sol discount is applied inside OpenRouter, on the OpenAI provider. OpenAI's own published price list still shows $5.00 per million input tokens and $30.00 per million output tokens for GPT-5.6 Sol, so calling OpenAI directly costs the full amount.

How long does the OpenRouter GPT-5.6 Sol discount last?

OpenRouter publishes no end date for the GPT-5.6 Sol discount. Its discounted-models page states only that provider discounts and promotional pricing can change, and that the collection updates as offers change. Anyone budgeting a long-running job around the lower rate should re-check the model page first.

How cheap does GPT-5.6 Sol get on the batch tier?

GPT-5.6 Sol on OpenRouter's batch tier costs $1.25 per million input tokens and $7.50 per million output tokens, half the already-discounted standard rate. Batch work that tolerates a delayed response is therefore the cheapest way to run Sol through OpenRouter.

Which other models are discounted alongside it?

OpenRouter's discounted-models collection lists GPT-5.6 Sol Pro at the same 50% cut, plus Google's Gemini 3.7 Flash at 50% off, DeepSeek V4 Flash at 44%, Z.ai GLM 5 at 40%, and steeper cuts on smaller open models such as inclusionAI Ling-2.6-flash at 90%.

OpenRouter · 2026-08-17 · major

GPT-5.6 Sol at half price — OpenRouter discounts OpenAI's flagship 50%

OpenRouter now sells GPT-5.6 Sol at 50% off: $2.50 per million input tokens and $15 per million output, against OpenAI's own list price of $5 and $30. The batch tier drops to $1.25 and $7.50.

OpenRouter page card for the GPT-5.6 Sol model

OpenAI's flagship model costs half as much through OpenRouter as it does buying from OpenAI directly.

Quick facts

Model	GPT-5.6 Sol
Discount	50% off list price
Where	OpenRouter only, OpenAI provider
Applies to	Standard, batch, and Sol Pro
Context window	1M tokens
Duration	Not stated; OpenRouter says offers can change

Pricing

Input · list $5.00	$2.50 / 1M tokens
Output · list $30.00	$15.00 / 1M tokens
Cached input · list $0.50	$0.25 / 1M tokens
Batch input	$1.25 / 1M tokens
Batch output	$7.50 / 1M tokens

source ↗

What is it?

A 50% discount on GPT-5.6 Sol is live on OpenRouter, the API marketplace that routes requests to model providers. Input falls from $5.00 to $2.50 per million tokens and output from $30.00 to $15.00. Cached input reads drop from $0.50 to $0.25, and GPT-5.6 Sol Pro carries the same 50% cut.

How does it work?

The discount sits on the OpenAI provider inside OpenRouter, so the model id openai/gpt-5.6-sol picks it up with no extra flag. Batch requests go lower still, at $1.25 input and $7.50 output per million tokens. OpenRouter lists the offer in its discounted-models collection, next to cuts on Gemini 3.7 Flash and several open models, and says promotional pricing can change.

Why does it matter?

Halving the token bill changes what is affordable to run on a frontier model: long agent loops, wide fan-out evaluation, and bulk document work on GPT-5.6 Sol all cost half as much through OpenRouter as calling OpenAI directly, where the list price is still $5.00 and $30.00 per million tokens. The 1M-token context window is the same on either path.

Who is it for?

teams running agents on frontier models

Frequently asked questions

Does the discount apply if I call OpenAI's API directly?: No. The GPT-5.6 Sol discount is applied inside OpenRouter, on the OpenAI provider. OpenAI's own published price list still shows $5.00 per million input tokens and $30.00 per million output tokens for GPT-5.6 Sol, so calling OpenAI directly costs the full amount.
How long does the OpenRouter GPT-5.6 Sol discount last?: OpenRouter publishes no end date for the GPT-5.6 Sol discount. Its discounted-models page states only that provider discounts and promotional pricing can change, and that the collection updates as offers change. Anyone budgeting a long-running job around the lower rate should re-check the model page first.
How cheap does GPT-5.6 Sol get on the batch tier?: GPT-5.6 Sol on OpenRouter's batch tier costs $1.25 per million input tokens and $7.50 per million output tokens, half the already-discounted standard rate. Batch work that tolerates a delayed response is therefore the cheapest way to run Sol through OpenRouter.
Which other models are discounted alongside it?: OpenRouter's discounted-models collection lists GPT-5.6 Sol Pro at the same 50% cut, plus Google's Gemini 3.7 Flash at 50% off, DeepSeek V4 Flash at 44%, Z.ai GLM 5 at 40%, and steeper cuts on smaller open models such as inclusionAI Ling-2.6-flash at 90%.

Try it

Call model id openai/gpt-5.6-sol on https://openrouter.ai