GPT-6.2 Sol for everyday agent workflows
Give GPT-6.2 Sol a ticket, a code excerpt or a screenshot and ask for a concrete next step. Current rates: $0.20 / 1M input, $10.00 / 1M output.
- Text and image input
- Structured output
- Responses API
- Chat Completions
Try GPT-6.2 Sol
Chat with GPT-6.2 Sol, attach screenshots, and see the tokens and cost of every answer.
gpt-6.2-sol$0.20 in · $10.00 out per 1M tokensGPT-6.2 Sol does not use temperature or Top P at any reasoning effort.
Try one of these prompts
Messages are billed from your balance at the rates below, like an API call. Failed requests are free.
Recorded structured output
This request used the Responses API with low reasoning effort and a strict JSON schema.
Extract the ticket owner and due weekday from: "Alex will send the corrected invoice by Friday."
{ "owner": "Alex", "due_weekday": "Friday" }
Generated on November 18, 2026. Whitespace is formatted for readability.
GPT-6.2 Sol pricing
GPT-6.2 Sol usage is billed in tokens. The table shows the current rates for each input length.
Example sizes are for illustration. Your bill follows the tokens each request reports.
| Category | Defaultgpt-6.2-sol | gpt 0.7gpt-6.2-sol:gpt-0.7 | ||
|---|---|---|---|---|
| ≤ 272,000 uncached input tokensPrice · 1M tokens | > 272,000 uncached input tokensPrice · 1M tokens | ≤ 272,000 uncached input tokensPrice · 1M tokens | > 272,000 uncached input tokensPrice · 1M tokens | |
| Input | $0.20 | $0.40 | $1.40 | $2.80 |
| Output | $10.00 | $1.50 | $70.00 | $10.50 |
| Cached input | $0.01 | $0.02 | $0.07 | $0.14 |
| Cache write | $0.25 | $0.50 | $1.75 | $3.50 |
Only uncached input determines the tier. Cached input is billed separately at the rate for that tier. Examples use the standard tier.
What is GPT-6.2 Sol?
GPT-6.2 Sol is an OpenAI reasoning model for everyday coding and repeatable agent tasks. Its 1,050,000-token context window includes the prompt and the answer. A request can supply up to 922,000 input tokens and generate up to 128,000 output tokens, including reasoning.
For a recurring task, define the result once. A JSON schema gives a ticket parser predictable fields, while a Responses function lets the model ask your application to look up an order.
gpt-6.2-sol · Text and image input · Text output · OpenAI Responses and Chat Completions
Two ways to use GPT-6.2 Sol
Call it from your code, or let your coding agent do it for you.
Call it from your backend
Use the OpenAI SDK with the SeedRouter base URL. Send gpt-6.2-sol in the model field.
- 1Create an API key
- 2Set the base URL to https://api.seedrouter.ai/v1
- 3Send gpt-6.2-sol as the model
Hand it to your coding agent
Copy a ready prompt. Your agent shows the request and cost before anything is spent.
- 1Copy the prompt
- 2Paste it into your agent
- 3Approve the request
What GPT-6.2 Sol is good at
Start with a repeatable input and define what an application should receive.
Extract typed records
Turn receipt lines into a list with a currency, an amount and a date for each purchase.
Call an application function
Supply a function definition in Responses. Validate the arguments before your application executes it.
Review a screenshot
Attach the image and ask for the visible error text and the steps a user can take.
What teams build with GPT-6.2 Sol
Support inbox
Convert each incident report into a category, an owner and an evidence excerpt.
Test fixtures
Generate an expected result from a small example and exercise it in your test suite.
Start using GPT-6.2 Sol in four steps
- 01
Create a key
Sign in and create an API key. Add credit whenever you need it; there is no subscription.
- 02
Change the base URL
Point the OpenAI SDK at https://api.seedrouter.ai/v1. Your code stays the same.
- 03
Pick GPT-6.2 Sol
Send gpt-6.2-sol as the model. Set reasoning effort from low to max to trade depth for speed and cost.
- 04
Check the bill
Every request shows its tokens and charge in your usage records.
GPT-6.2 Sol on SeedRouter at a glance
What you can use today.
- OpenAI Responses
- Yes, official format
- OpenAI Chat Completions
- Yes
- Streaming
- Yes
- Function calling
- Use the Responses API
The model produces text. Send a screenshot as image input when an answer depends on what is visible.
GPT-6.2 Sol limits
1.05M-token context
On GPT-6.2 Sol, prompt, images and history share one 1.05M-token window; input is capped at 922K.
128K tokens out
One GPT-6.2 Sol request can return up to 128K tokens, reasoning included.
Long prompts cost more
Use the input threshold shown in the pricing table to select the rates for the request.
No sampling controls
GPT-6.2 Sol does not use temperature or top_p at any reasoning effort.
Why call GPT-6.2 Sol through SeedRouter
OpenAI SDK support
Use your existing OpenAI client with the SeedRouter base URL and API key.
Pay per token
Top up once and spend it on any model. No plan, no monthly fee.
Failures are free
A request that errors is not charged.
Live prices
The rates on this page are the rates your account pays.
Prompt caching
Place stable source material first. Send new user turns after that prefix. Read the usage object to find the number of input tokens served from cache.
Image input
Send screenshots, charts and photos by URL with your prompt.
Streaming
Stream GPT-6.2 Sol tokens as they arrive with stream: true.
Related models
Other models you can call with the same key.
Call GPT-6 Astra with the OpenAI SDK. OpenAI's most capable model, with a 1.05M-token context, live per-token pricing and a playground you can try right now.
View pricingCall GPT-6 Luna with the OpenAI SDK. The lowest-cost GPT-6 model, with a 1.05M-token context, live per-token pricing and a playground you can try right now.
View pricingCall GPT-6 Sol with the OpenAI SDK. 1.05M-token context, reasoning effort from none to max, live per-token pricing and a playground you can try right now.
View pricingCall GPT-6.1 Sol with the OpenAI SDK: near-Astra coding and agent work at a fifth of the price, 1.05M-token context, 95% off cached input. Try it now.
View pricingGPT-6.2 Astra is OpenAI’s flagship model for demanding reasoning and coding. It can build and revise software and work across documents in a single project.
View pricingCall Claude Fable 5 with the Anthropic or OpenAI SDK. Deep reasoning, 1M-token context, forced tool use, live per-token pricing and a playground to test it.
View pricingGPT-6.2 Sol guides
Walkthroughs and comparisons for this model.
GPT-6.2 Sol FAQ
What is GPT-6.2 Sol?
GPT-6.2 Sol is an OpenAI reasoning model for everyday coding and repeatable agent tasks. Its 1,050,000-token context window includes the prompt and the answer. A request can supply up to 922,000 input tokens and generate up to 128,000 output tokens, including reasoning.
Which endpoint should I use for tools with GPT-6.2 Sol?
Use /v1/responses and put function definitions in tools. Read the function_call item in output. Your application runs the function and returns a function_call_output item with the matching call_id.
How much does GPT-6.2 Sol cost?
Input is billed at $0.20 / 1M and output at $10.00 / 1M. Use the pricing table for cache and long-context rates.
What reasoning efforts does GPT-6.2 Sol support?
For new integrations OpenAI recommends starting at medium. Choose low for shorter steps. In Responses set reasoning.effort. In Chat Completions set reasoning_effort. Both models reject none.
What is the knowledge cutoff of GPT-6.2 Sol?
OpenAI lists June 1, 2026 as the knowledge cutoff for GPT-6.2 Sol. Include newer source material in your prompt when freshness matters.
What is the GPT-6.2 Sol model ID and how do I call it?
The model ID is gpt-6.2-sol. Create a SeedRouter API key, point the OpenAI SDK at https://api.seedrouter.ai/v1 and send gpt-6.2-sol as the model with the Responses or Chat Completions API.
Can GPT-6.2 Sol read images?
Yes. GPT-6.2 Sol accepts images alongside text, sent by URL in the request. It returns text only.
Am I charged if a GPT-6.2 Sol request fails?
No. A GPT-6.2 Sol request that returns an error is not charged. You pay only for the tokens a finished request reports.
Try GPT-6.2 Sol now
Send your first GPT-6.2 Sol request from the playground or your own code in minutes.
