GPT-6 Sol pricing is $2.00/1M input and $10.00/1M output per 1M tokens on the standard tier. Cached input is $0.200/1M. Cache writes are $2.50/1M. These are permanent list rates, about half of GPT-5.6 Sol. Rates in TokenCalculator checked 23 Sep 2026.
Price your own prompt
Sticker rates are only the start. Paste a real prompt into the cost calculator with GPT-6 Sol selected, then set output size, cache hit rate, and monthly volume.
GPT-6 Sol rate card
API model id: gpt-6-sol. Context window: about 1.05M tokens. Max output: 128K. Exact OpenAI encodings (o200k_base) apply in the TokenCalculator tokenizer.
| Line | USD per 1M tokens |
|---|---|
| Input (<=272K prompt) | $2.00/1M |
| Cached input (<=272K) | $0.200/1M |
| Cache write | $2.50/1M |
| Output (<=272K) | $10.00/1M |
| Batch input | $1.00/1M |
| Batch output | $5.00/1M |
| Long-context input (>272K, whole request) | $4.00/1M |
| Long-context output (>272K, whole request) | $15.00/1M |
| Long-context cached input (>272K) | $0.400/1M |
Permanent pricing, not a launch promo
OpenAI frames GPT-6 Sol as permanent standard rates at roughly half GPT-5.6 Sol ($4.00/1M / $20.00/1M). Plan migrations on that cut, then re-check quality. Do not assume ChatGPT Work or Codex credits match API $/1M.
The 272K long-context cliff
When prompt tokens cross 272K, OpenAI reprices the entire request, not only the overflow. Input and cache rates double. Output rises to 1.5x of the standard band. Design agents around that threshold before you treat the full 1.05M window as a flat price.
Fast mode and Batch
Batch and Flex are about half of standard rates when latency can wait. Fast mode is about 2x standard for lower latency. EU data residency uses Standard processing only. Confirm the service tier on OpenAI’s pricing page before you ship a production path.
How Sol compares to nearby tiers
| Model | Input / 1M | Output / 1M | Cache read / 1M |
|---|---|---|---|
| GPT-6 Sol | $2.00/1M | $10.00/1M | $0.200/1M |
| GPT-6 Luna | $0.100/1M | $0.500/1M | $0.010/1M |
| GPT-6 Astra | $10.00/1M | $50.00/1M | $1.00/1M |
| GPT-5.6 Sol | $4.00/1M | $20.00/1M | $0.400/1M |
| Claude Opus 5.5 | $4.00/1M | $20.00/1M | $0.200/1M |
| Grok 4.7 (<200K) | $2.00/1M | $6.00/1M | $0.500/1M |
Sol sits between Luna and Astra on OpenAI. Versus Claude Opus 5.5, Sol is half on input and output stickers while cache reads tie at $0.20 per 1M. Versus Grok 4.7, Sol matches input under 200K but Grok output is cheaper on the card. Cost per finished task can still flip if one model burns more tokens.
Worked examples
- Short chat: 2K fresh input + 500 output about $0.009 on standard rates.
- Cached agent turn: 20K cached + 2K fresh + 1K output about $0.018 (cache read $0.20/M).
- Long prompt over 272K: use the long-context band for the whole request, not only the tokens past the threshold.
When to use GPT-6 Sol
Use Sol when GPT-6 Luna fails your evals on coding or agent jobs and you do not need Astra list rates. Prefer Luna for high-volume chat and routing. Prefer Astra when Sol still fails hard end-to-end work. Prefer Opus 5.5 when Anthropic tooling or Claude quality wins your harness.
Common mistakes
- Shopping only on the $2 / $10 sticker and ignoring cache, Fast mode, and the 272K cliff
- Treating ChatGPT plan credits as API $/1M
- Comparing Sol to Opus 5.5 on list price alone without equalizing output and cache mix
- Budgeting Claude with Sol Exact token counts
- Treating curated TokenCalculator rates as a live OpenAI price websocket
Frequently asked questions
How much does GPT-6 Sol cost?
Standard API pricing is $2.00/1M per 1M input tokens and $10.00/1M per 1M output tokens. Cached input is $0.200/1M. Confirm on OpenAI’s pricing page for your account tier.
Is GPT-6 Sol pricing permanent or promotional?
TokenCalculator catalogs it as permanent standard rates at about half GPT-5.6 Sol. Confirm on OpenAI pricing for your account. Do not assume temporary launch discounts.
What is the GPT-6 Sol API model id?
Use gpt-6-sol in the OpenAI API.
Does Sol have a long-context surcharge?
Yes. Prompts over 272K input tokens reprice the full request at higher input, cache, and output rates.
Is Fast mode more expensive?
Yes. Fast mode is about 2x standard rates for lower latency. Batch and Flex are about half when delayed processing is acceptable.
How does Sol compare to Claude Opus 5.5 on price?
Sol lists at $2.00/1M / $10.00/1M. Opus 5.5 lists at $4.00/1M / $20.00/1M. Cache reads both sit at $0.20 per 1M. Read the Sol vs Opus 5.5 guide for workload math.
Is GPT-6 Sol cheaper than GPT-6 Astra?
Yes on list rates. Astra is $10.00/1M / $50.00/1M. Use Astra when Sol quality is not enough.
Should I migrate from GPT-5.6 Sol now?
GPT-6 Sol lists at half GPT-5.6 Sol stickers in this catalog. Re-run evals, then switch when quality holds. Price the same prompt on both in the cost calculator.
Are these rates live?
No. TokenCalculator curates official list prices and stamps lastVerified (23 Sep 2026). Confirm critical budgets on the provider page.
Where should I start in TokenCalculator?
Open the cost calculator with GPT-6 Sol preselected, paste a representative prompt, set output tokens and monthly volume, then compare Luna, Astra, and Opus 5.5 on the same text.
Next steps
- GPT-6 Sol vs Claude Opus 5.5
- GPT-6 Luna pricing
- GPT-6 Astra pricing
- OpenAI API pricing hub
- How we verify rates
- Official OpenAI pricing
Related Sep 2026 pricing guides
Frontier rate cards, cost-per-task math, and long-context cliffs. Cross-link these when you compare Sol, Luna, Opus 5.5, Grok 4.7, or Astra.