Grok 4.5 API
Choose Grok 4.5
An evaluation route for difficult reasoning, long-context analysis, research, and workflows that combine text generation with xAI server-side tools.
Grok 4.5
xAI reasoning and tool-use model
grok-4.5Research with web and X search, code-assisted analysis, document and collection retrieval, complex reasoning, and provider-diverse production routing.
Grok 4.5 pricing
Estimate token and successful server-side tool usage together. Live prices for your user group override the contract-backed fallback rates.
Request calculator
Enter uncached input, cached input, output, and successful tool calls.Estimated request cost
Grok 4.5At 200,000 prompt tokens (uncached + cached input), every token role is billed at 2×. Minimum total charge: 0.01 credits. Tool calls are not multiplied.
Budget guide
Approximate requests using the current mix.For quick testing
For regular development
For production evaluation
Token rates
| Model | Prompt tokens | Uncached input tokens | Cached input tokens | Output tokens |
|---|---|---|---|---|
Grok 4.5grok-4.5 | < 200K | $1.700 / 1M-15% 115.6 cr / 1M$2.000xAI official | $0.256 / 1M-15% 17.4 cr / 1M$0.300xAI official | $5.100 / 1M-15% 346.8 cr / 1M$6.000xAI official |
Grok 4.5grok-4.5 | ≥ 200K | $3.400 / 1M-15% 231.2 cr / 1M$4.000xAI official | $0.512 / 1M-15% 34.8 cr / 1M$0.600xAI official | $10.200 / 1M-15% 693.6 cr / 1M$12.000xAI official |
USD and credits are shown per 1M tokens. The long-context tier starts inclusively at 200,000 prompt tokens and applies 2× to input, cached input, and output.
Responses server-side tool rates
Only successful server-side calls on /v1/responses are billed per call. Ordinary client function calls and failed attempts have no server-tool fee; tokens still apply.
What is the Grok 4.5 API?
Grok 4.5 is xAI’s reasoning model for coding agents, connected research, long-document analysis, and tool-driven workflows. The page URL uses grok-4-5, while API requests use the model ID grok-4.5. Grok 4.5 is live on the EvoLink production route; review pricing boundaries, supported workflows, and model configuration in the Pricing and API sections below.
What is the Grok 4.5 API best suited for?
Grok 4.5 combines a 500K-token context window, configurable reasoning, image input, and tools for work that must retain evidence, call external systems, and return verifiable results. These use cases explain where it may add value and what to confirm before integration.
Repository-scale coding and code review
Keep relevant source files, issues, test results, and previous changes in one task for cross-file debugging, implementation planning, and review. Before rollout, use fixed repository tasks to check test pass rate, incomplete steps, structured results, and required human edits instead of judging coding ability from a single demo.
Long documents and multi-source analysis
The 500K-token context can hold reports, contracts, knowledge-base passages, conversation history, and retrieved sources together. More context does not automatically improve an answer, so verify that key evidence is preserved and use the long-context and cached-input rates in Pricing to calculate cost per acceptable result.
Connected research and tool workflows
Grok 4.5 can combine web search, X search, code execution, attachment search, and collections search into sourced research. When using EvoLink, check tool availability, token usage, and per-call charges in the existing API and Pricing sections, then test citations and result quality on your actual research flow.
Structured output and agent orchestration
Text and image inputs can feed JSON Schema, function calling, and multi-step agent flows for extraction, review, and downstream automation. Before production use, test schema validity, function arguments, streaming completion events, and safe recovery after a tool failure.
What changes when you access Grok 4.5 through different platforms?
Even with the same model, platforms can differ in context configuration, tool support, usage reporting, and billing. EvoLink puts model configuration and usage behind one API so later model changes remain straightforward.
Use the correct API model ID
grok-4-5 is the page URL and a common search form; the API request model ID is grok-4.5. Existing Chat Completions or Responses applications can use the correct ID in model configuration, with exact fields available in the API section.
Follow the active API route configuration
xAI documents a 500K-token model window, but platforms may expose different context settings, tools, and rate limits. When using EvoLink, rely on the model configuration, available features, and actual usage shown for the current route.
Choose Chat or Responses by workflow
Start with Chat Completions for standard chat, streaming, and client-side functions. Choose Responses when you need research agents, server-side search, or code execution. This preserves a familiar OpenAI-style integration without adding complexity the workload does not need.
Keep model choice open in one gateway
Use one EvoLink account, balance, and API pattern for Grok, GPT, Claude, and Kimi. Keeping model selection in configuration lets teams route by task quality, cost, and availability without rebuilding application code for every provider.
How can you control Grok 4.5 API cost more accurately?
The existing Pricing section shows current token and tool rates. In practice, caching, context management, reasoning settings, and model routing help reduce unnecessary usage and connect spend to completed business tasks.
Make repeated context cache-friendly
Stable system prompts, tool schemas, and shared context are easier to reuse through caching. Configure the supported cache or conversation identifier for the selected protocol and inspect cached tokens in usage to confirm that repeated requests receive the expected benefit.
Send only the context the task needs
A 500K-token window is useful for large repositories and documents, but it does not need to be filled on every request. Selecting only relevant files, messages, and retrieved passages reduces input cost and helps the model focus on the evidence that matters.
Tune reasoning, output, and tool calls by task
Start simple work with lower reasoning effort and shorter output, then increase the reasoning budget for harder analysis. Research and agent workflows should also track tool-call count to avoid repeated searches, executions, or unproductive loops.
Compare models by total task cost
Evaluate tokens, cached input, server tools, and required retries within the same completed task. EvoLink centralizes model and usage information so teams can compare the total cost of completing equivalent work across Grok 4.5 and other routes.
How should you choose Chat Completions, Responses, and production traffic?
The two protocols serve different workflows. This is a selection summary; use the existing API section and EvoLink documentation for exact request fields, and the existing Pricing section for token and tool charges.
Standard chat and client functions: Chat Completions
If you already use OpenAI-compatible chat, streaming, or client-side function calling, start with Chat Completions. Confirm message format, streaming completion, function arguments, and usage against what the current client expects.
Research agents and server tools: Responses
Evaluate Responses when you need web or X search, code execution, attachment or collections search, and longer multi-step agent flows. Confirm tool availability, call charges, citations, failure states, and retry boundaries before integration.
Large-context work: watch cache and total cost together
Long documents, repositories, and conversations should not automatically put everything into one request. Compare representative context sizes, cache hits and misses, output length, and retries, then use the final charge to calculate successful-task cost.
Production traffic: start small and keep a fallback
Send a small, observable set of tasks to Grok 4.5 first and retain a proven GPT, Claude, or Kimi route. EvoLink’s unified API centralizes model choice, usage, and balance, making it easier to change routes after rate limits, timeouts, schema errors, or tool failures.
What should you confirm before using Grok 4.5 in production?
Start with a small set of real workloads to determine whether the model meets your quality, latency, cost, and reliability needs before deciding which traffic should move.
Integration and usage data are clear
Confirm that the application uses the correct grok-4.5 model ID and intended protocol, and that required responses, usage, and cache information are returned. Clear usage data supports cost analysis and gives Chat and Responses workflows a consistent observation method.
Outputs meet real business requirements
Test real code changes, long-document analysis, research, or structured extraction. Beyond answer quality, check whether tests pass, citations are reliable, function arguments are correct, and JSON Schema output can be consumed directly by downstream systems.
Latency and error handling meet expectations
Observe response time under representative traffic and prepare retry behavior for rate limits, timeouts, invalid structured output, and tool failures. Configurable model selection in EvoLink makes it easier to switch to a verified alternative when one route is temporarily unavailable.
Cost and model choice remain controllable
Use Pricing, usage, and final charges to calculate total cost for the same class of task, then decide whether Grok 4.5 belongs on the default route, difficult tasks, or fallback traffic. A unified gateway keeps quality and budget decisions separate from integration work.
Begin with a small, observable, reversible set of Grok 4.5 tasks. Expand only after quality, latency, and cost meet expectations. Keeping multiple model options behind the EvoLink unified API makes later scaling, switching, and cost optimization easier.
API model ID
grok-4.5grok-4-5 is the page URL. Requests on Chat Completions and Responses must use the exact model ID shown here.
Related models
GPT-5.6
OpenAI’s tiered frontier family for comparing capability, latency, and cost-routing flexibility.
View model
Claude Opus 5
Anthropic’s premium route for long-running agents, tool use, and complex review.
View model
DeepSeek V4
DeepSeek’s cost-sensitive open-model route for high-volume coding, reasoning, and agent workloads.
View modelRelated guides
Grok 4.6 vs Grok 4.5
Review the evidence and rollout gates before treating a reported successor as a production upgrade.
Read guide
Grok 4.6 release status
Track confirmed release and API evidence without turning the Grok 4.5 product page into a rumor or news page.
Read guideGrok 4.5 API FAQ
Is the Grok 4.5 API available through EvoLink now?
Yes. Grok 4.5 is live on the EvoLink production route. Send requests with the model ID grok-4.5 on Chat Completions or Responses, and check the model ID, pricing, context, and supported workflows on this page before scaling up.
Is grok-4-5 the API model ID?
No. grok-4-5 is the page slug and search label. The actual request model ID is grok-4.5, shown in the model ID section on this page.
What is the Grok 4.5 context window and long-context pricing boundary?
xAI documents a 500,000-token context window. Requests enter the long-context price tier at the prompt-token threshold shown in Pricing, so test representative context sizes and cached input instead of assuming every request should use the full window.
What inputs and outputs does Grok 4.5 support?
Upstream documentation lists text and image input with text output. Image input means the model can analyze visual information; it does not make Grok 4.5 an image-generation model. Confirm the effective EvoLink route limits during validation.
Should I use Chat Completions or the Responses API?
Use Chat Completions for familiar chat, streaming, and client function flows. Evaluate Responses for longer-running agent workflows and server-side tools. The current API section and EvoLink docs remain the source for exact request fields and verified route support.
How should reasoning effort be selected?
Start with the lowest setting that meets the task quality requirement, then compare medium and high on the same tasks. Complex work may benefit from a larger reasoning budget, while simple work can prioritize response time and cost.
How do cached input and long agent loops affect Grok 4.5 pricing?
Cached input can reduce repeated-context cost, while cache misses, long outputs, retries, and repeated tool steps can make an agent loop more expensive than the headline input rate suggests. Use the live Pricing section and final request charges when calculating cost per accepted task.
What is the difference between server tools and client functions?
Client functions run in your application. Web and X search, code execution, attachment search, and collections search are xAI server tools. They differ in execution location, failure handling, and billing; check the current support and rates in the API and Pricing sections.
How should teams compare Grok 4.5 with GPT, Claude, or Kimi?
Run the same real tasks with consistent context, tools, and reasoning settings. Compare result quality, response time, token mix, tool performance, and total cost to decide which EvoLink route fits each traffic class.
What fallback should teams keep during rollout?
Keep a model that already handles the same workload reliably and leave route selection configurable. If rate limits, timeouts, or invalid output occur, EvoLink can route to GPT, Claude, Kimi, or another suitable alternative.