GPT-5.6 Sol pricing, context window, and cost calculator
An OpenAI model with tracked short-context, long-context, cached-input, and cache-write pricing.
Verified fields
Quick facts
gpt-5.6-solSupported capabilities
- Streaming
- Function calling
- Structured outputs
- Prompt caching
- Batch API
- Reasoning
Only capabilities explicitly tracked from provider documentation are shown.
Source-tracked rates
Verified pricing profiles
Amounts are USD per 1M tokens unless the column states an hourly storage unit.
| Profile | Input / 1M | Cached input / 1M | Cache write / 1M | Output / 1M | Notes |
|---|---|---|---|---|---|
| standard / short contextDefault | $4.00 USD | $0.40 USD | $5.00 USD | $20.00 USD | Used for default estimates. |
| standard / long context | $8.00 USD | $0.80 USD | $10.00 USD | $30.00 USD | Tracked separately; select explicitly when modeling this profile. |
Official price history
GPT-5.6 Sol price changes
Verified changes for the same model and pricing profile. Cross-model generational price differences are not treated as history.
OpenAI · Standard · Short Context
GPT-5.6 Sol
| Rate | Earlier price | Later recorded price | Change | Percentage |
|---|---|---|---|---|
| Input / 1M | $5.00 | $4.00 | -$1.00 | -20.0% |
| Cached input / 1M | $0.50 | $0.40 | -$0.10 | -20.0% |
| Output / 1M | $30.00 | $20.00 | -$10.00 | -33.3% |
| Cache write / 1M | $6.25 | $5.00 | -$1.25 | -20.0% |
Price change recorded from the current official provider pricing source.
OpenAI · Standard · Long Context
GPT-5.6 Sol
| Rate | Earlier price | Later recorded price | Change | Percentage |
|---|---|---|---|---|
| Input / 1M | $10.00 | $8.00 | -$2.00 | -20.0% |
| Cached input / 1M | $1.00 | $0.80 | -$0.20 | -20.0% |
| Output / 1M | $45.00 | $30.00 | -$15.00 | -33.3% |
| Cache write / 1M | $12.50 | $10.00 | -$2.50 | -20.0% |
Price change recorded from the current official provider pricing source.
Workload estimate
GPT-5.6 Sol cost calculator
Change the workload assumptions. The calculator uses the same shared pricing resolver as StackLens Compare.
Estimated monthly total$140.00
- Monthly input cost
- $40.00
- Monthly output cost
- $100.00
- Per 1,000 requests
- $14.00
- Active profile
- standard / short context
Estimate excludes untracked provider-specific charges.
StackLens cost guidance
When GPT-5.6 Sol switches to long-context pricing
GPT-5.6 Sol uses a higher tracked rate when a request exceeds 272,000 input tokens. These single-request examples show the pricing change on either side of that boundary.
Below the long-context threshold
- Input
- 250,000 tokens · $1.00
- Output
- 10,000 tokens · $0.20
- Estimated token cost
- $1.20
At 250,000 input tokens, the shared resolver keeps the request on the tracked short-context profile.
Above the long-context threshold
- Input
- 300,000 tokens · $2.40
- Output
- 10,000 tokens · $0.30
- Estimated token cost
- $2.70
At 300,000 input tokens, the tracked long-context input and output rates apply to the request.
Boundary: The threshold example shows token pricing, not model quality, latency, retries, or the value of using a larger context window.
Compare GPT-5.6 Sol with Kimi K3 →Compare GPT-5.6 Sol with Grok 4.5 →
Explicit assumptions
Common workload examples
1M input tokens
$4.00 under the standard.short_context profile, excluding output and other charges.
1M output tokens
$20.00 under the standard.short_context profile, excluding input and other charges.
10,000 monthly requests
At 1,000 input and 500 output tokens per request: $140.00 under standard / short context.
50% cached input
The same default workload with 50% cached input is estimated at $122.00. Cache savings apply only to the tracked cached-input rate.
Research layer
Experience and rollout notes
Official facts are separated from user reports. Community reports are useful signals, not controlled benchmarks.
Official long-context pricing boundary
OpenAI documents a 1.05M-token context window and 128K maximum output for GPT-5.6 Sol. Requests above 272K input tokens use a higher published input and output pricing tier.
What this does not prove: The pricing threshold is an official billing rule, not a statement about quality at long context.
Planning and code review
Early users repeatedly highlight planning, code review, and concise explanations as useful strengths in development workflows.
What this does not prove: These reports do not use a shared task set or scoring method.
Implementation scope and usage
A recurring counter-signal is that Sol may overbuild straightforward implementations or consume more tokens than expected, which can offset gains on simpler tasks.
What this does not prove: Measure completed-task cost and review time on your own workload rather than relying on token price alone.
No controlled StackLens benchmark yet
StackLens has not run a controlled GPT-5.6 Sol benchmark. The notes above separate official documentation from independent user reports.
Research reviewed 2026-07-13. Reports may change as GPT-5.6 Sol reaches more workflows.
Tracked relationships
Related models
StackLens relationship map based on model families and published decision comparisons. Find a workload match → Open relationship map → Open price map →
Featured comparisons
- GPT-5.6 Sol vs GPT-6 Astra cost
- GPT-5.6 Sol vs Kimi K3 cost
- GPT-5.6 Sol vs Kimi K2.7 Code High-Speed cost
- GPT-5.6 Sol vs Kimi K2.7 Code cost
- GPT-5.6 Sol vs Grok 4.5 cost
- GPT-5.6 Sol vs GPT-5.6 Terra cost
Continue your research
Verification
Sources and methodology
Pricing and limits are source-tracked and may change. Verify current values with the provider before making production purchasing decisions.
Calculator profile rule: standard.short_context is the verified default. Requests with more than 272,000 input tokens use the higher long-context rate for the full request.