gpt-6-sol API — per-token pricing and code examples
Price per 1M tokens
Cache rates apply when the provider reports cache tokens.
87% offAbout this model
gpt-6-sol is served through OpenAI Chat Completions, OpenAI Responses, Anthropic Messages. You pay $0.26 per 1M input tokens and $1.30 per 1M output tokens. That is 87% below the official price of $2.00 and $10.00. Cache reads cost $0.044 per 1M tokens when the provider reports cached tokens. It is included in budget packs from $9.99 up.
What GPT-6 Sol is built for
OpenAI says GPT-6 Sol is built to power complex coding and agentic workflows. It came out on September 22, 2026, the same day as GPT-6 Luna, and OpenAI's changelog describes the pair as reasoning models that accept text and image inputs and generate text through the Responses and Chat Completions APIs. The Sol model page now points to GPT-6.1 Sol as the newer model, so GPT-6 Sol makes most sense when you already tuned prompts and evaluations against it and want stable behaviour while you compare.
On paper the limits are generous: a 1,050,000-token context window with up to 922,000 tokens of input, and up to 128,000 tokens of output in one response. That is enough to put a large slice of a repository, a long design document and the agent's own history into a single request. OpenAI gives April 20, 2026 as its knowledge cutoff.
Effort levels and the Chat Completions tool rule
The GPT-6 Sol model page lists effort levels none, low, medium, high, xhigh and max, with medium as the default. For coding agents the interesting range is usually medium and above, while none suits quick helper calls such as naming a branch or summarising a commit.
The detail that matters most for agent builders: OpenAI's page says function calling through Chat Completions on GPT-6 Sol works only with reasoning_effort set to none. OpenAI documents tool calling with reasoning for GPT-6 Sol on its Responses API, along with its built-in tools for this model, such as web search, file search, code interpreter, MCP and apply patch. On Cent AI, the /v1/responses format is available for Sol.
When effort is anything other than none, OpenAI asks you to remove temperature, top_p and top_logprobs, and on Chat Completions logprobs too. A coding agent that still sets temperature: 0 from an older setup will need that line removed.
{
"model": "gpt-6-sol",
"reasoning": { "effort": "high" },
"input": "Find the race condition in this worker and propose a patch"
}
The example uses the /v1/responses format. On /v1/chat/completions the same choice is the top-level reasoning_effort field.
What it costs on Cent AI
Every GPT-6 Sol request is charged per token from your prepaid USD balance, with the four rates shown in the price box on this page. Reasoning tokens are counted by OpenAI under usage.output_tokens_details.reasoning_tokens and billed as output. In a long agent loop at high or xhigh, that hidden thinking can be the largest part of the bill, so compare a few runs at medium before you settle on a level.
OpenAI lists prompt caching as supported for GPT-6 Sol. The cache read and cache write rates are applied only to the cached tokens that come back in OpenAI's usage report. Cent AI never assumes a hit. Agent loops that resend the same system prompt and tool list each turn tend to benefit the most.
Using it from your tools
You can call GPT-6 Sol on /v1/chat/completions, /v1/responses and /v1/messages. Cent AI does not change your parameters silently, so a parameter GPT-6 Sol does not accept comes back as an upstream_error with the model's status code, for example 400; Cent AI does not drop it for you. If a request ends with status: "incomplete" and the reason max_output_tokens, raise the output limit; OpenAI recommends leaving roughly 25,000 tokens for reasoning plus the answer.
Specification
- Context window
- —
- Input and output
- Text
- Endpoints
- /v1/chat/completions, /v1/responses, /v1/messages
- Smallest pack
- $9.99
Latency and error rate per model are on the status page. status.cent-ai.dev
What a request costs
Open the full calculatorCost per request
Estimate
gpt-6-solAn estimate at price version 5. Your bill comes from the usage the provider reports for each request.
Save $0.3982Code
curl https://api.cent-ai.dev/v1/chat/completions \
-H "Authorization: Bearer $CENT_AI_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-6-sol", "messages": [{"role": "user", "content": "Hello"}]}'Set CENT_AI_KEY to a key from the dashboard. OpenAI SDKs use https://api.cent-ai.dev/v1; Anthropic SDKs use https://api.cent-ai.dev.
Price history
Version 5, in effect since Nov 12, 2026.
Effective
per 1M tokens
Inputold $0.26 new $0.26Cache readold $0.026 new $0.044Cache writeold $0.26 new $0.26Outputold $1.30 new $1.30Effective
per 1M tokens
Inputold $0.26 new $0.26Cache readold $0.026 new $0.026Cache writeold $0.325 new $0.26Outputold $1.30 new $1.30Effective
per 1M tokens
Inputold $0.10 new $0.26Cache readold $0.01 new $0.026Cache writeold $0.10 new $0.325Outputold $0.50 new $1.30Effective
per 1M tokens
Inputold $0.10 new $0.10Cache readold $0.01 new $0.01Cache writeold $0.125 new $0.10Outputold $0.50 new $0.50
Compare with
- hy4-previewInput$0.46704
- deepseek-v4-proInput$0.165
- gpt-6.1-solInput$0.58
- gpt-6.1-terraInput$0.15
Questions
How much does gpt-6-sol cost on Cent AI?
$0.26 per 1M input tokens and $1.30 per 1M output tokens. Cache reads cost $0.044 and cache writes $0.26 per 1M tokens. Each request is rounded up once.
Is there a minimum charge?
A successful request costs at least $0.00002. A request that ends early, because the client left or the output was cut off, pays only for its tokens.
How do I pay?
Top up a prepaid USD balance with QRIS or USDT/USDC, or buy a budget pack once. Nothing renews automatically.
What if the provider does not report usage?
The request is billed from an estimate and marked estimated on the Requests page. An estimate never contains cache tokens.
Which endpoint do I call?
/v1/chat/completions, /v1/responses, /v1/messages. The OpenAI base URL is https://api.cent-ai.dev/v1.
Try gpt-6-sol with one key
One prepaid balance for every model. Pay with QRIS or USDT/USDC.