Grok 4.5 Review: Pricing, Performance, and How It Compares to Claude and GPT
Grok 4.5 is xAI's coding-focused AI model, released July 8, 2026. It's built into Cursor (all plans) and the xAI API, priced at $2/$6 per million input/output tokens. For everyday consumer use, xAI's bundled X Subscription (from $8/month, launched October 12, 2026) replaced the older SuperGrok lineup; see the update below.
xAI's coding-focused model, built after acquiring Cursor and priced well below Claude and GPT. Here's how it performs and where it actually fits.
Grok 4.5 Review: Pricing, Performance, and How It Compares to Claude and GPT
Grok 4.5 is xAI's coding-focused model, first released on July 8, 2026 as the company's first launch since going public and acquiring the AI coding editor Cursor, aimed squarely at developers rather than casual chatbot users. This review covers what it is, how it performs, what it costs against Claude and GPT, how to access it, and who should try it.
Update (August 2026): Grok 4.6 Has Launched
Just over a month after Grok 4.5 shipped, xAI released Grok 4.6 on August 12, 2026, extending the same coding-and-agent focus with a longer supplemental training run, a new "xhigh" reasoning tier for harder problems, and the ability to stream its reasoning summaries as it works, something Grok 4.5 does not expose.
Headline pricing stayed the same at $2 per million input tokens and $6 per million output tokens for prompts under 200K tokens (crossing that line bills the whole request at $4/$12, same as before). The one quiet change: cached-input pricing rose from $0.30 to $0.50 per million tokens. On the Artificial Analysis Intelligence Index, Grok 4.6 scored 61, five points above Grok 4.5, though it used roughly 47% more output tokens to get there, pushing the real cost of a standard task from $0.36 to $0.84, more than double.
In short: Grok 4.6 is measurably smarter but meaningfully more expensive to run at scale. Some price trackers now list an even newer Grok 4.7 at the same per-token rates, so check xAI's own documentation for the current model list. Everything below about Grok 4.5's pricing, performance, and access still applies if you are on that model.
What Grok 4.5 Actually Is
Unlike earlier Grok releases built as general-purpose chat assistants, Grok 4.5 was designed from the ground up for coding, agentic tasks, and knowledge work. xAI trained it using real developer session data pulled from Cursor, which the company acquired earlier in 2026, meaning the model learned from how people actually write and debug code in a live editor, not just from scraped code repositories.
Elon Musk described it as an "Opus-class" model, positioning it as a direct rival to Anthropic's higher-end coding models, but with a specific emphasis on speed and cost rather than raw benchmark supremacy.
How It Performs
On independent benchmarking, Grok 4.5 lands in a solid but not top-tier position: it ranks fourth on the Artificial Analysis Intelligence Index, ahead of every Gemini model and open-weight competitor, though behind Anthropic's top coding model and OpenAI's GPT-5.5 on several engineering benchmarks. Where it does stand out is efficiency: xAI reports it uses roughly four times fewer output tokens than Claude's Opus 4.8 to solve the same engineering tasks, and it leads on at least one real-world benchmark measuring sustained, multi-step coding work.
Speed and cost
Grok 4.5 runs at around 80 tokens per second, with xAI signaling that a custom inference stack in development could roughly double that speed without changing the underlying model.
Pricing: Where It Gets Interesting
This is the part most developers actually care about. Here are the current list prices for Grok 4.5 and its main rivals:
| Model | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|
| Grok 4.5 | $2 | $6 |
| Claude Opus 5 | $5 | $25 |
| GPT-5.6 Sol | $4 | $20 |
| GPT-5.6 Luna | $0.20 | $1.20 |
At those rates, Grok 4.5 undercuts the flagship models from Anthropic and OpenAI by a wide margin on output cost, while OpenAI's budget Luna tier is cheaper still. Prices here reflect recent cuts: GPT-5.6 Sol was reduced from $5/$30 on August 21, and Luna from $1/$6 on July 30. For teams running high-volume agentic workflows where token costs add up fast, the difference matters.
What that means on a real workload
Say an agentic coding workflow sends 10 million input tokens and 2 million output tokens a month. At list prices that costs about $32 on Grok 4.5 (10 × $2 + 2 × $6), about $80 on GPT-5.6 Sol, about $100 on Claude Opus 5, and about $4.40 on GPT-5.6 Luna. Cost is not the only factor, because a model that needs more retries or produces weaker code can cost more in practice than its sticker price suggests, so test on your own tasks.
Where to Access Grok 4.5
Grok 4.5 is available as the default model in Grok Build, inside Cursor on all plans, and through the xAI API console. Cursor's own pricing page now lists plans alongside Grok, with the lower tiers centered on Grok models, so Cursor has become a natural place to try it.
SuperGrok: Consumer Access and Pricing
Update, October 12, 2026: xAI launched The X Subscription, one bundle covering Grok, Grok Bot, Cursor and X Premium: Free, Plus ($8), Premium ($30), Super ($100) and Ultra ($200) per month. According to Cursor's pricing page, Premium and above include the latest Grok models, while Plus runs Grok models only. The SuperGrok tiers below are the lineup from before that launch, and xAI says existing subscribers can keep them. Our Cursor pricing guide explains the new plans in detail.
The API pricing above is for developers building with Grok 4.5 directly. For everyday use through grok.com or the X app, access runs through xAI's consumer subscription tiers instead:
- Free: limited access to Grok models
- SuperGrok Lite ($10/month): entry-level paid tier with higher limits
- SuperGrok (~$30/month): raises usage limits and unlocks DeepSearch and Big Brain mode
- SuperGrok Heavy ($300/month, discounted to $99/month for the first three months): the only consumer plan with confirmed full Grok 4.5 access as of late July 2026
xAI also offers Grok access bundled through X: X Premium ($8/month) and X Premium+ ($40/month). Consumer plans in this area have been changing quickly, so treat these as a snapshot from launch time and confirm current plans before you subscribe. If your goal is specifically to use Grok 4.5 itself rather than earlier Grok models, SuperGrok Heavy was the tier that guaranteed it outside of Cursor and the API.
Who Should Use Grok 4.5
If your work leans heavily on agentic coding, meaning long, multi-step tasks where an AI has to plan, write, test, and fix code across several rounds, Grok 4.5 is a legitimate option to test alongside Claude and GPT-based tools, particularly if cost per task is a real constraint. It is not the outright top performer on every benchmark, but the combination of speed, price, and native Cursor integration makes it worth a trial run.
A simple way to trial it
- Pick five real tasks from your backlog, such as a bug fix, a refactor, and a small feature.
- Run each through Grok 4.5 and your current model with the same instructions.
- Compare how many attempts each needed, the quality of the result, and the total tokens used.
- Decide on cost per finished task, not cost per token.
For more on how the leading models compare, read our Claude vs ChatGPT comparison, or browse the AI tools directory.
Frequently Asked Questions
When was Grok 4.5 released?
Grok 4.5 launched on July 8, 2026 — xAI's first model release since going public and acquiring Cursor.
What is SuperGrok, and do I need it to use Grok 4.5?
SuperGrok was xAI's consumer subscription line, separate from the API pricing above. On October 12, 2026 xAI launched The X Subscription (Free, Plus $8, Premium $30, Super $100, Ultra $200), where Premium and above include the latest Grok models. Developers can skip consumer plans entirely and use Grok 4.5 through Cursor or the xAI API instead.
Is Grok 4.5 better than Claude for coding?
It depends on the task. Independent benchmarks show Anthropic's top coding model still leads on most engineering tests, but Grok 4.5 is significantly cheaper and more token-efficient, and it wins on at least one real-world sustained-coding benchmark.
Can I use Grok 4.5 in Cursor?
Yes — it's available in Cursor on all plans, alongside xAI's own Grok Build tool.
Has Grok 4.6 replaced Grok 4.5?
Not exactly -- Grok 4.6 launched August 12, 2026 as a newer, more capable model at the same headline price, but Grok 4.5 remains available. If cost per task matters more than squeezing out the last few points of benchmark performance, Grok 4.5 is still a reasonable pick; Grok 4.6 is the better choice when you need the extra capability and can absorb the higher effective cost. Some trackers now list a newer Grok 4.7 as well, so check xAI's documentation for the current lineup.
Was Grok 4.5 available in the EU at launch?
Not immediately — xAI rolled it out first in Grok Build, Cursor, and the API console, with EU availability following in mid-July 2026, about a week after the initial launch.
Is Grok 4.5 a general chatbot or a coding-specific tool?
It's built and trained primarily for coding, agentic tasks, and knowledge work, though it can still handle general questions like other Grok models.
How much cheaper is Grok 4.5 than Claude or GPT?
At list prices of $2 input and $6 output per million tokens, a workload of 10 million input and 2 million output tokens costs about $32 on Grok 4.5, versus about $100 on Claude Opus 5 and about $80 on GPT-5.6 Sol. OpenAI's budget Luna tier is cheaper still at about $4.40, but the best choice depends on how many retries each model needs.
User Reviews
Be the first to review!