Grok 4.5: xAI's 500K-Context Model

Grok 4.5 is xAI's current general-purpose model on magicdoor.ai. It accepts text and image input, supports reasoning and function calling, and can work with as much as 500K tokens of context.

That makes it a useful option for long conversations, large text inputs, image analysis, and tasks that benefit from reasoning or tool use. The two details to plan around are its tiered long-context pricing and the fact that PDF input and code interpreter are not enabled for Grok on magicdoor.ai.

Open the Grok 4.5 model page for the current model summary.

Grok 4.5 Capabilities on magicdoor.ai

Image input means Grok can analyze an image included in the conversation. Image generation is a separate workflow: magicdoor.ai can route that part of the task to one of its supported image models, then let you continue the conversation without managing a separate provider account.

Grok 4.5 Pricing

Grok pricing depends on the prompt size for each request:

The higher tier starts at exactly 200K prompt tokens, not after 200K. Because the threshold applies per request, a month containing many short prompts still uses the lower tier even if their combined token count exceeds 200K.

Cost examples

• A request with 100K input tokens and 20K output tokens costs about $0.32: $0.20 input plus $0.12 output.
• If all 100K input tokens in that request are served from cache, the same input portion costs $0.03 instead of $0.20.
• A request with 250K input tokens and 50K output tokens costs about $1.60 at the long-context rates: $1.00 input plus $0.60 output.

These are model-usage costs. magicdoor.ai has a $6 monthly base subscription that includes $1 in credits, with additional usage billed from your pay-as-you-go balance.

When Grok 4.5 Is a Good Fit

Choose Grok 4.5 when the task needs one or more of these:

• a large text context or a long-running conversation
• analysis of text and images in the same prompt
• reasoning before producing a final answer
• tools exposed through function calling
• a second model's perspective inside a multi-model workflow

The 500K window is a capacity limit, not a reason to include unnecessary material. Shorter, focused context is usually cheaper and easier to verify. Once a single prompt reaches 200K tokens, both input and output move to the higher pricing tier.

When to Switch Models

Grok 4.5 is not the right model for every turn. Switch to a supported OpenAI model when you need PDF input or code interpreter. Use Magicdoor's search integration when the answer depends on current web information, and use an image model when the task requires generating or editing an image rather than analyzing one.

Because magicdoor.ai preserves the conversation while you switch models, one practical workflow is:

1. Use Grok 4.5 to reason over text or image input.
2. Switch models for PDF processing, code execution, search, or image generation when needed.
3. Return to Grok 4.5 or another chat model for synthesis.
4. Watch the live cost display, especially before sending a 200K-token prompt.

Prompting Grok 4.5

Grok 4.5 does not need a special prompt syntax. Give it the same ingredients you would give another capable model:

• the outcome you need
• the relevant source material
• constraints and non-goals
• the desired format
• a request to identify uncertainty or missing evidence

For a long-context review, ask the model to cite the section, heading, or excerpt supporting each conclusion. For image analysis, say what to inspect rather than asking for a generic description. For reasoning tasks, define the decision criteria before asking for a recommendation.

FAQ

How much does Grok 4.5 cost on magicdoor.ai?

Grok 4.5 costs $2 per 1 million input tokens and $6 per 1 million output tokens when the prompt is below 200K tokens. At 200K prompt tokens or more, it costs $4 for input and $12 for output. Cached input costs $0.30 below 200K or $0.60 at 200K and above.

What can Grok 4.5 handle?

Grok 4.5 accepts text and image input, supports reasoning and function calling, and has a 500K-token context window. On magicdoor.ai, it can also participate in search, image, and canvas workflows through the platform's existing integrations.

Does Grok 4.5 support PDFs or code interpreter on magicdoor.ai?

No. Grok 4.5 does not currently support PDF input or code interpreter on magicdoor.ai. Switch to a supported OpenAI model when those tools are required.

The Bottom Line

Grok 4.5 combines a 500K context window, text and image input, reasoning, and function calling with a lower base token price than the model it replaces. Its best use on magicdoor.ai is inside a flexible workflow: use Grok where its context and capabilities fit, then switch models when the task needs PDFs, code interpreter, search, or image generation.