Drafted with AI tools and edited by a human. The figures and conclusions were checked by a GPUniq editor.
TL;DR
GPT-6 Sol has a 1,050,000 token context window and a 128,000 token max output. Standard pricing: $2.00 per million input tokens, $10.00 per million output tokens. Long-context rates: $4.00 input, $15.00 output. Cached input: $0.20 per million, cache writes: $2.50 per million. OpenAI reports about 50 percent fewer coding errors than GPT-5.6 Sol.
GPT-6 Sol API pricing
Standard input: $2.00 per million tokens. Output: $10.00 per million. Long-context (over ~272,000 input tokens) costs $4.00 per million input, $15.00 per million output. Cached input: $0.20 per million. Cache writes: $2.50 per million. All pricing is from OpenAI's API documentation.
| Token Type | Standard $/1M | Long-context $/1M | Cached $/1M |
|---|---|---|---|
| Input | 2.00 | 4.00 | 0.20 |
| Output | 10.00 | 15.00 | N/A |
| Cache Writes | 2.50 | N/A | N/A |
Long-context rates only apply over ~272,000 input tokens. Caching is cheap and useful for repeated inputs.
Context window size
GPT-6 Sol supports a 1,050,000 token context window. That is enough for huge codebases or multi-document processing in one API call.
Token limits
Max output: 128,000 tokens. Max input: 1,050,000 tokens (entire context window). You can generate or analyze massive amounts of text or code in a single run.
Coding error rate
OpenAI's internal benchmark shows GPT-6 Sol makes about 50 percent fewer coding errors than GPT-5.6 Sol on their coding tasks.
| Model | Coding Error Rate (relative) |
|---|---|
| GPT-5.6 Sol | 1.0x |
| GPT-6 Sol | 0.5x |
Source: OpenAI model comparison.
Reasoning effort levels
GPT-6 Sol supports these reasoning effort levels: none, low, medium, high, xhigh, max. You can pick the effort level to balance cost and accuracy per task.
Access and availability
- Available via OpenAI API as model id "gpt-6-sol"
- In ChatGPT Work and Codex for Plus, Pro, Business, Enterprise, and Edu users
- Not in the general Chat interface for Free/Go plans
- Free/Go users can access Luna (not Sol) in the desktop app
See OpenAI's official documentation for updates.
Caching and long-context rates
Long-context input: $4.00 per million tokens. Long-context output: $15.00 per million. Standard rates for smaller contexts.
Cached input: $0.20 per million. Cache writes: $2.50 per million.
| Feature | Standard $/1M | Long-context $/1M | Cached $/1M |
|---|---|---|---|
| Input | 2.00 | 4.00 | 0.20 |
| Output | 10.00 | 15.00 | N/A |
| Cache Writes | 2.50 | N/A | N/A |
If you process repeated inputs, use caching. For large projects, expect long-context rates.
Related on the GPUniq blog
Methodology
Facts are compiled from the vendor's own announcement and documentation, linked below, at the time of writing; where the vendor hasn't published a number it is marked as such. GPUniq prices, when the model is already available here, come from our live catalog.
Sources
- 1.GPT-6 Sol Model | OpenAI API — OpenAI (accessed )
- 2.Changelog | OpenAI API — OpenAI (accessed )
- 3.Pricing | OpenAI API — OpenAI (accessed )
Want cheap GPUs for your next project?
Browse live GPU prices and rent the right card in seconds — H100, A100, RTX 4090, and 50+ more models.
