GPT-6.2 Sol, GPT-6.1 Sol and GPT-6.1 Terra

The newest GPT-6 models are here. GPT-6.2 Sol, GPT-6.1 Sol and GPT-6.1 Terra from OpenAI are in the model picker starting today, and two Codex models join them for longer coding work.
What you can do with them
- Take on hard problems. The Sol models always reason before they answer, so they suit research, analysis and tricky code.
- Work with big inputs. The GPT-6 models have a context window of 1.05 million tokens, so a long codebase, a set of reports or a whole chat history fits in one conversation.
- Ask about a picture. Attach a screenshot, a chart or a photo of a page and ask what it shows.
- Look things up. Web search is built in, for questions about what is happening now.
- Choose how hard it thinks. Pick low, medium or high when a problem needs more care.
Which one to pick
- GPT-6.2 Sol is the newest of the three. Start here for complex work.
- GPT-6.1 Sol is the previous Sol, still available alongside 6.2.
- GPT-6.1 Terra is the lighter one, for capable everyday work. It is the only one of the three where you can turn reasoning off for quicker replies.
All three start at 1 premium request per send, on paid plans. You can change models in the middle of a chat, so try the same prompt on two of them and compare.
Codex for longer coding work
GPT-5.2-Codex and GPT-5.1-Codex-Max are OpenAI coding models, built for longer coding sessions with many steps. Both have a 400,000 token context window, let you choose low, medium or high reasoning, and start at 1 premium request per send, on Premium+ plans.
Where they run, and what happens to your chats
On just4o.chat, the GPT-6 models are served by OpenAI. The two Codex models are served on Microsoft Azure.
Requests are sent through the Vercel AI Gateway with zero data retention, and your chats are not used for training.
Your conversation history still lives in your just4o.chat account, the same as with every other model, so you can come back to it later.
At a glance
| GPT-6.2 Sol | GPT-6.1 Sol | GPT-6.1 Terra | |
|---|---|---|---|
| Cost per send | From 1 premium request | From 1 premium request | From 1 premium request |
| Image input | Yes | Yes | Yes |
| Reasoning | Low, medium or high | Low, medium or high | None, low, medium or high |
| Web search | Yes | Yes | Yes |
| Context window | 1.05 million tokens | 1.05 million tokens | 1.05 million tokens |
| Privacy level | Zero data retention | Zero data retention | Zero data retention |
| Plans | Paid plans | Paid plans | Paid plans |
| GPT-5.2-Codex | GPT-5.1-Codex-Max | |
|---|---|---|
| Cost per send | From 1 premium request | From 1 premium request |
| Reasoning | Low, medium or high | Low, medium or high |
| Context window | 400,000 tokens | 400,000 tokens |
| Privacy level | Zero data retention | Zero data retention |
| Plans | Premium+ plans | Premium+ plans |
Longer chats and higher reasoning settings use more requests on any of them.
Try it
Open chat, choose GPT-6.2 Sol in the model picker, and give it the hardest question on your list. Full details are on the GPT-6.2 Sol, GPT-6.1 Sol, GPT-6.1 Terra, GPT-5.2-Codex and GPT-5.1-Codex-Max model pages.

