GPT-6.2 Sol: OpenAI's Everyday Model Gets Better at the Same Price
By the ChatGBot Team · Published November 18, 2026
OpenAI released GPT-6.2 Sol on November 17, 2026, next to the flagship GPT-6.2 Astra. Sol is the model OpenAI wants you to use every day: for coding, writing and agents at scale. It gets clearly better than GPT-6.1 Sol and keeps the same price. Here is what changed, how it scores and when to pick it over Astra.
Every top AI in one app
GPT, Claude, DeepSeek, Qwen and GLM. One subscription.
Try ChatGBot →GPT-6.2 Sol at a glance
| GPT-6.2 Sol | |
|---|---|
| Made by | OpenAI |
| Released | November 17, 2026 |
| API model ID | gpt-6.2-sol |
| Replaces | GPT-6.1 Sol |
| Context window | 1,050,000 tokens |
| Max output | 128,000 tokens |
| Input | Text and images |
| Output | Text |
| Knowledge cutoff | June 2026 |
| API price per 1M tokens | $2 input / $10 output |
| In ChatGPT | ChatGPT Work and Codex (Plus, Pro, Enterprise, Edu) |
What is GPT-6.2 Sol?
OpenAI calls Sol its go-to model for everyday work, and says many of its own staff use it that way. It sits below Astra in power but far below it in price: Sol costs a quarter of Astra per input and output token. Pro 200 users also get the Ultrafast speed mode for the first time with GPT-6.2 Sol.
What is new
- Same price, better results. $2 and $10 per million tokens, unchanged from GPT-6.1 Sol.
- Near-Astra coding. OpenAI says Sol is within 0.5 points of Astra on FrontierCode 2.0 at about one fifth of the cost per task.
- Beats last month's flagship. On LifeVal it beats GPT-6 Astra at about one seventh of the cost.
- Better at work tasks. 5.3 points higher than GPT-6.1 Sol on GDPval v2.
Benchmarks
| Test (OpenAI, max reasoning) | GPT-6.1 Sol | GPT-6.2 Sol |
|---|---|---|
| Agents' Last Exam v1.1 | 72.5% | 73.0% |
| Terminal-Bench-Science 0.1 | 57.0% | 68.0% |
| FrontierCode 2.0 | 39.0% | 43.8% |
| LifeVal | 79.5% | 83.8% |
Outside tests: Sol scores 57 on the Artificial Analysis Intelligence Index. Artificial Analysis also measured a 55% hallucination rate on hard factual questions and coding scores 3 points below GPT-6.1 Sol, so the picture is not all positive.
GPT-6.2 Sol vs GPT-6.2 Astra
| GPT-6.2 Sol | GPT-6.2 Astra | |
|---|---|---|
| Best for | Everyday coding, writing and agents | The hardest, longest projects |
| API price per 1M tokens | $2 / $10 | $8 / $40 |
| FrontierCode 2.0 | 43.8% | 44.2% |
| Agents' Last Exam v1.1 | 73.0% | 77.6% |
| Artificial Analysis index | 57 | 62 |
For most tasks, Sol gives you most of Astra's quality for a quarter of the price. Move up to Astra when a job is long, high stakes or scientific.
Safety
Like Astra, Sol is rated Critical for cybersecurity and High for biology and chemistry in OpenAI's system card, and it comes with the same tighter limits on risky security help. Unlike Astra, it does not reach High for AI self-improvement.
Price and how to get access
| Per 1M tokens | Input | Cached input | Output |
|---|---|---|---|
| GPT-6.2 Sol | $2 | $0.10 | $10 |
| GPT-6.1 Sol | $2 | $0.10 | $10 |
| GPT-6.2 Sol, long context (over 272K) | $4 | $0.20 | $15 |
| GPT-6.2 Sol, Batch | $1 | $0.05 | $5 |
- ChatGPT: ChatGPT Work and Codex for Plus, Pro, Enterprise and Edu.
- API: model ID gpt-6.2-sol.
- GPT-6.1 Sol is still listed in the API at the same price.
Should you use it?
If you already use GPT-6.1 Sol, switching is an easy upgrade at no extra cost. For writing fiction and long prose, some users still prefer GPT-6.1 Terra. Trying the same prompt on a few models is the quickest way to know which one fits your work. See also our November 2026 AI model releases roundup.
Use top models in one place
Ready to try it yourself?
Switch between top AI models in one chat.
Try ChatGBot →Frequently asked questions
What is GPT-6.2 Sol?
GPT-6.2 Sol is OpenAI's everyday model for coding, writing and agents, released on November 17, 2026. It replaces GPT-6.1 Sol at the same price.
How much does GPT-6.2 Sol cost?
In the API it costs $2 per million input tokens and $10 per million output tokens, the same as GPT-6.1 Sol. Cached input costs $0.10 per million tokens.
Is GPT-6.2 Sol as good as GPT-6.2 Astra?
Close on some tasks. OpenAI says Sol is within 0.5 points of Astra on FrontierCode 2.0 at about one fifth of the cost per task. Astra is still stronger on the hardest work.
Who can use GPT-6.2 Sol?
Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex, and developers in the API as gpt-6.2-sol.
What is the GPT-6.2 Sol context window?
1,050,000 tokens, with up to 128,000 output tokens. It reads text and images and answers in text.
Can I use GPT models in ChatGBot?
Yes. ChatGBot offers GPT, Claude, DeepSeek, Qwen and GLM models in one app, so you can compare answers side by side.