What Grok is known for
Grok models are direct, quick to answer, and come with some of the largest context windows available. The table shows which ones read images and reason before replying.
Grok models · xAI
xAI's Grok models without an X account or a Premium plan. Same chat, same memory, same history as every other model on Portail.
Every xAI chat model Portail carries today, listed below.
The largest context window in the family, for long files and long chats.
Switch from Grok to GPT, Claude, Gemini or anything else mid-conversation.
Grok models are direct, quick to answer, and come with some of the largest context windows available. The table shows which ones read images and reason before replying.
Pick a Grok model from the model picker in any chat. It reads the whole conversation, including anything other models said earlier, and your Portail memory. When you want a second opinion, switch models and keep going, or turn on Council to ask several models at once.
Every plan, Free included, can use every Grok model. Plans change the size of your monthly credit allowance, not which models you can pick. Bigger models use credits faster; the cost band in the table shows roughly how fast.
Read from Portail's live model catalog. Cost runs from $ (cheapest) to $$$$.
| Model | Context | Images | Reasoning | Cost |
|---|---|---|---|---|
| Grok Build 0.1 | 256K | Yes | Yes | $$ |
| Grok 4.20 Multi-Agent | 2M | Yes | Yes | $$ |
| Grok 4.20 | 2M | Yes | Yes | $$ |
| Grok 4.6 | 500K | Yes | Yes | $$$ |
| Grok 4.5 | 500K | Yes | Yes | $$$ |
| Grok 4.3 | 1M | Yes | Yes | $$ |
6 right now, from xAI. The list follows the live catalog, so new Grok models appear as they're added.
Yes. All 6 Grok models on Portail accept images.
Yes. The Free plan includes every model, Grok too, with a starter credit allowance and no card. Cheaper Grok models stretch it further.
No. Grok on Portail runs on your Portail account and credits. X features, like Grok reading live posts on X, stay on X.
Yes, two ways. Switch the model and ask again in the same thread, or turn on Council to send one message to several models and get the best answer picked for you.
Every model, one thread, one memory. Free to start.