GPT-5.6 Sol
by OpenAI · openai/gpt-5.6-sol · #303 cheapest of 325 paid models
Prices updated Jul 22, 2026, 1:31 AM UTC · refreshed hourly
GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly strong at command-line and multi-step coding tasks...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5Gemini 3.6 Flash
Pricing
Input
$5.00
per 1M tokens
Output
$30.00
per 1M tokens
Blended (3:1)
$11.25
3 input : 1 output
Cache read
$0.50
per 1M cached tokens
Cache write
$6.25
per 1M tokens
Cache & batch economics
Effective prices for GPT-5.6 Sol with prompt caching.
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $11.25 | $0.54 | — |
| 50% | $9.56 | $0.315 | −42% |
| 90% | $8.21 | $0.135 | −75% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates for this provider’s batch API. Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.0115 | $11.50 |
| Document summary | 8,000 | 1,000 | $0.07 | $70.00 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.21 | $210.00 |
| Long-context analysis | 150,000 | 5,000 | $0.9 | $900.00 |
Performance
Intelligence Index
58.9
AA composite quality
Coding Index
77.4
Math Index
—
Agentic Index
54
Output speed
70 tok/s
median
Time to first token
112.90s
median
Time to first answer
112.90s
after reasoning tokens
Artificial Analysis benchmarks
| Benchmark | Score |
|---|---|
| GPQA Diamond | 94.1% |
| Humanity's Last Exam | 47.2% |
| SciCode | 56.1% |
| IFBench | 72.7% |
| AA-LCR (Long Context Reasoning) | 73.7% |
| Terminal-Bench Hard | 65.9% |
| Terminal-Bench 2.1 | 88.0% |
| τ²-Bench (Telecom) | 85.1% |
| τ³-Bench Banking | 33.0% |
Specs
Context window
1.05M
tokens
Max output
128K
tokens
Input modalities
file, image, text
Tokenizer
GPT