OpenAI
PreviewGPT-5.6 Sol pricing
The top tier of OpenAI's June 2026 GPT-5.6 family, tuned for the hardest problems: complex coding and security research.
Input
$5.00/ 1M tok
Output
$30.00/ 1M tok
- Context window
- 1.5M
- Max output
- 128K
- Cached input
- $0.5000 / 1M
- Verified
- 2026-07-01
GPT-5.6 family
| Tier | Input / 1M | Output / 1M | Cached in / 1M |
|---|---|---|---|
Sol (this page) Hardest problems: complex coding, security research | $5.00 | $30.00 | $0.5000 |
Terra High-volume business tasks | $2.50 | $15.00 | $0.2500 |
Luna Fast everyday work | $1.00 | $6.00 | $0.1000 |
Sol is the flagship tier. See the Terra and Luna pages for the mid and low tiers. GPT-5.6 launched in limited preview on 26 June 2026 (~20 organisations).
GPT-5.6 Sol is the top of OpenAI's three-model GPT-5.6 family, released 26 June 2026. It is aimed at the hardest problems – complex coding and security research – and is priced accordingly at $5 per 1M input tokens and $30 per 1M output tokens, the same rate card as the earlier GPT-5.5 flagship.
GPT-5.6 introduces more predictable prompt caching: cache reads receive the standard 90% discount (so Sol's cached input is $0.50 / 1M), while cache writes bill at 1.25x the uncached input rate, with explicit cache breakpoints and a 30-minute minimum cache life. Context is reported at roughly 1.5M tokens, up from 1M on GPT-5.5, though OpenAI has not published an official figure.
Because the family spans a 5x price range from Luna to Sol, reserve Sol for genuinely hard requests and route latency-tolerant, high-volume work to Terra or Luna. As of launch the models are limited to a small preview cohort.
Estimate LLM costs before you send
Paste your prompt into the Calcis estimator to see token counts and per-request cost across every tracked model, then compare GPT-5.6 Sol against them side by side.
Frequently asked
- How much does GPT-5.6 Sol cost per 1M tokens?
- $5 for input and $30 for output per 1M tokens. Cached input reads bill at $0.50 / 1M (a 90% discount).
- When should I use Sol instead of Terra or Luna?
- Sol is tuned for the hardest problems - complex coding and security research. For high-volume business tasks use Terra ($2.50/$15), and for fast everyday work use Luna ($1/$6).
- What is the GPT-5.6 Sol context window?
- Reported at approximately 1.5M tokens. OpenAI has not published an official figure at launch.
- How does prompt caching work on GPT-5.6 Sol?
- Cache reads get the standard 90% discount ($0.50 / 1M input); cache writes bill at 1.25x the uncached input rate. GPT-5.6 adds explicit cache breakpoints and a 30-minute minimum cache life.
- Is GPT-5.6 Sol generally available?
- Not at launch. It debuted 26 June 2026 for a limited preview cohort of roughly 20 organisations, with wider availability expected in the following weeks.
Pricing verified 2026-07-01 from the provider's rate card. These figures are informational and not yet wired into the Calcis estimator or billing.