On July 8, 2026, xAI released Grok 4.5, its first model built specifically for coding and agentic work. This is not a general capability upgrade. It is a deliberate pivot toward developers, with training done using real Cursor developer session data and benchmarks designed to measure what an AI can actually do inside a real codebase over a long session.
The result is a model that lands fourth on the Artificial Analysis Intelligence Index, above every open-weight model and notably above all Gemini models, at a price over 60% lower than Claude Opus 4.8 or GPT-5.5. For coding workloads specifically, independent benchmarks put it on par with GPT-5.5 in Codex at roughly half the per-task cost.
Elon Musk positioned it directly: “It is an Opus-class model, but faster, more token-efficient and lower cost.” In a follow-up he was more precise: “Grok 4.5 is roughly comparable to Opus 4.7, but much faster.”
Musk has also flagged that current speeds are not the ceiling. xAI has not yet deployed its internally developed C/C++ inference stack that maps directly to the GB300 hardware, and when it does, he expects speeds to double or more. A further step-change release is signaled for next month.
The Key Takeaways
- Grok 4.5 launched July 8, 2026 and ranks 4th on the Artificial Analysis Intelligence Index (score 54), ahead of every Gemini and open-weight model.
- Built on the 1.5-trillion-parameter V9 foundation and trained on real Cursor session data for coding and agentic tasks.
- Priced at $2 / $6 per million tokens, over 60% cheaper than Opus 4.8 and GPT-5.5, at roughly half the per-task cost of GPT-5.5 in Codex.
- Highly token-efficient, using about 14,000 output tokens per Intelligence Index task versus 67,020 for Opus 4.8.
- Musk calls it “Opus-class, but faster,” with a custom inference stack expected to double the current ~80 tokens/sec.
What Grok 4.5 Actually Is
Grok 4.5 is xAI’s first model trained from the ground up for coding and agentic tasks rather than general intelligence. Built on the 1.5-trillion-parameter V9 foundation model, it was trained in partnership with Cursor to handle long-running jobs across multiple repositories and to operate with minimal human intervention across hundreds of tool calls.
The headline specs are straightforward.
| Spec | Detail |
|---|---|
| API model name | grok-4.5 |
| Architecture | V9 foundation model, 1.5 trillion parameters (3x larger than Grok 4.3’s V8) |
| Context window | 500k tokens |
| Speed | approximately 80 tokens per second |
| Input modalities | text and images (vision) |
| Reasoning | configurable at low, medium, or high effort (default high) |
| Supported APIs | Responses API and Chat Completions |
| Built-in capabilities | function calling, web search, X search, code execution |
The V9 name marks a full generation shift from the V8-series that powered Grok 4.3, tripling the scale while shifting the training focus entirely toward coding and agentic tasks. The context window dropped from Grok 4.3’s 1 million tokens to 500k, which is a real tradeoff for anyone who needed the full million.
The Cursor Connection
The “trained with Cursor” framing in the announcement is not just a marketing partnership. In mid-June 2026, SpaceX acquired Cursor for approximately $60 billion, bringing one of the most widely used AI coding editors directly into the xAI ecosystem. That acquisition is why Grok 4.5’s training incorporated real Cursor developer session data, giving xAI direct access to the actual workflows, context patterns, and long-horizon task structures that developers use in production.
Unlike training on public code repositories, real session data captures how developers iterate on a problem, context-switch across files, and make multi-step decisions inside actual projects. Grok 4.5 is the first model to reflect that.
xAI also introduced a new training methodology alongside this model, asynchronous learning, which allows multi-hour agentic training runs to proceed in parallel with ongoing model training rather than sequentially. The practical effect is that feedback loops between the model’s behavior and its training updates are much tighter, which is what enables the model to handle the kind of long-running autonomous sessions that most agents currently stall on.
Benchmark Results
xAI’s Published Benchmarks
| Benchmark | Fable 5 | GPT-5.5 | Grok 4.5 | Opus 4.8 | GLM-5.2 |
|---|---|---|---|---|---|
| DeepSWE 1.0 | 66.1% | 64.3% | 62.0% | 55.8% | n/a |
| DeepSWE 1.1 | 70.0% | 67.0% | 53.0% | 59.0% | 44.0% |
| Terminal-Bench 2.1 | 84.3% | 83.4% | 83.3% | 78.9% | n/a |
| SWE-Bench Pro | 80.4% | n/a | 64.7% | 69.2% | 62.1% |
Reading this straight: Grok 4.5 beats Opus 4.8 on two of these four benchmarks (DeepSWE 1.0 and Terminal-Bench 2.1) and trails it on two others (DeepSWE 1.1 and SWE-Bench Pro). Claude Fable 5 leads all four. On Terminal-Bench 2.1, the margin between Grok 4.5 (83.3%) and GPT-5.5 (83.4%) is a single tenth of a point.
Independent Benchmarks from Artificial Analysis
On the Artificial Analysis Intelligence Index, Grok 4.5 at high reasoning scores 54 and ranks fourth out of 168 models. It trails only Claude Fable 5, GPT-5.5, and Claude Opus 4.8. That is a 16-point improvement over Grok 4.3, the single largest generation-over-generation jump xAI has posted on this index. It outperforms every open-weight model and every Gemini model.
On GDPval-AA v2, the benchmark measuring sustained agentic knowledge work, Grok 4.5 ranks fourth with an Elo of 1,543. It sits between Claude Opus 4.8 (Elo 1,600) and GLM-5.2 (Elo 1,513). On the toughest financial reasoning sub-task, tau3-Banking, Grok 4.5 hits 33%, the top score among all models evaluated, beating GPT-5.5 (xhigh) at 31%.
On the Artificial Analysis Coding Agent Index (DeepSWE, Terminal-Bench v2, and SWE-Atlas QnA combined), Grok 4.5 in the Grok Build harness scores 76, ranking third. It is on par with GPT-5.5 (xhigh) in Codex and sits just below Fable 5 (max) in Claude Code.
xAI also reports the Harvey Legal Agent Benchmark, where Grok 4.5 ranks first. Combined with the tau3-Banking top score, that points at legal and financial knowledge work as specific strengths beyond what the coding benchmarks alone capture.
Token Efficiency
The per-token price is the headline, but token efficiency is the more meaningful number. On SWE-Bench Pro, Grok 4.5 used approximately 15,954 output tokens per task versus Opus 4.8’s 67,020, a 4.2x gap. On the Artificial Analysis Intelligence Index it uses roughly 14,000 output tokens per task, over 60% fewer than Opus 4.8. On the Coding Agent Index it averages 1.9 million total tokens per task, versus 7.2 million for Fable 5 in Claude Code and 6.2 million for GPT-5.5 in Codex.
A model that uses 4x fewer tokens is not just cheaper per task. It is faster and produces less noise in agentic pipelines, which matters for multi-step autonomous work.
Pricing and Cost Per Task
Grok 4.5 is priced at $2.00 per million input tokens and $6.00 per million output tokens. Cache hits are discounted 75%, bringing cached input tokens to $0.50 per million. One important caveat, pricing doubles for inputs longer than 200k tokens. For a full breakdown of Grok subscription and API pricing, including the consumer tiers, that guide covers all the options.
The per-token price is over 60% lower than Opus 4.8 and GPT-5.5. But cost per task is the more meaningful number, since it accounts for how many tokens each model actually spends.
| Model | Input (per 1M) | Output (per 1M) | Cost per Intelligence Index task | Cost per Coding Agent task |
|---|---|---|---|---|
| Grok 4.5 | $2.00 | $6.00 | $0.31 | $2.49 |
| GPT-5.5 (Codex) | $5.00 | $30.00 | higher | $5.07 |
| Fable 5 (Claude Code) | $10.00 | $50.00 | higher | $11.80 |
| Opus 4.8 | $5.00 | $25.00 | higher | n/a |
The per-task figures are what actually matter at production scale. Grok 4.5 achieves near-equivalent coding agent performance at roughly half the cost of GPT-5.5 and less than a quarter the cost of Fable 5 per completed job.
Where Grok 4.5 Is Available
Grok 4.5 went live across a wide set of platforms on launch day. It is available in the xAI API under the model name grok-4.5, in Grok Build (xAI’s own coding agent harness), and in Cursor across all plans. Third-party access covers OpenRouter, Vercel, Cloudflare, Snowflake, and Databricks Mosaic. Fello AI support is expected within the next few days.
The Office plugin launch is noteworthy. xAI is positioning Grok 4.5 not just as a coding model but as a knowledge work model capable of constructing complex Excel models with integrated web research and generating sophisticated PowerPoint content. That broadens the addressable use case well beyond software development.
One availability gap: Grok 4.5 is not yet available in the EU, either through xAI’s products or the API console. xAI has indicated EU availability is targeted for mid-July 2026.
The model supports both the Responses API and Chat Completions formats, meaning it slots into codebases already built on OpenAI-compatible APIs without modification.
What Elon Musk Said
Musk made several statements around launch day that point at where this model fits and where it is going.
On capability positioning: “It is an Opus-class model, but faster, more token-efficient and lower cost,” with a follow-up clarification that it is “roughly comparable to Opus 4.7, but much faster.” That last detail matters. Musk is benchmarking against Opus 4.7, not 4.8, which the independent data broadly supports.
On the speed ceiling: “Grok 4.5 is not yet using our internally developed C/C++ inference software that exact maps to the GB300 hardware. Doubling or more of the current speed is probably achievable.” Current throughput around 80 tokens per second is not the final figure. The custom inference stack is in development and will add further speed without changing the model weights.
On what is next: “Next month’s release will be another step-change improvement, as we close the loop on solving real-world engineering problems at Tesla, SpaceX, Neuralink and Boring Company.” That feedback loop from real engineering work at xAI’s sister companies is a training data source no other lab has at the same scale.
On future multimodal capabilities: Grok will gain the ability to call Grok Imagine as a tool in agentic mode, invoking image and video generation as part of a longer autonomous workflow rather than as a separate feature. Musk flagged this as especially valuable for game developers.
Early User Reactions
Developer feedback from early Cursor access was positive. Danny Limanseta described it as “Opus 4.8 at 2x the speed at a much cheaper price point” after using it to brainstorm, plan, and implement a complex game feature across a full session without needing to manually correct it at each step.
Artificial Analysis summarized the competitive shift bluntly: Grok 4.5 “brings SpaceXAI to the intelligence frontier behind only OpenAI and Anthropic, and outperforming all open weights models and notably Google’s Gemini models.”
How Grok 4.5 Compares to the Competition
Against the current frontier, Grok 4.5 rarely takes the outright top spot on raw capability, but it consistently wins on cost and token efficiency. Here is how it stacks up model by model.
Versus Grok 4.3
A 16-point Intelligence Index jump is the biggest generational leap in the Grok lineage. The context window halved from 1M to 500k, which is the one real tradeoff. If you have been using Grok 4.3 for coding, Grok 4.5 is a significant upgrade across every other dimension.
Versus Claude Opus 4.8
Grok 4.5 beats Opus on DeepSWE 1.0 and Terminal-Bench 2.1 but trails on DeepSWE 1.1 and SWE-Bench Pro. On the Intelligence Index and GDPval-AA v2, Opus still leads. The per-task cost comparison heavily favors Grok 4.5. For price-sensitive agentic workloads, Grok 4.5 makes a real case. For the hardest open-ended coding tasks, Opus still has an edge.
Versus Claude Fable 5
Fable 5 leads across all four xAI benchmarks and on the Intelligence Index. The cost gap is the largest of any comparison: $2.49 versus $11.80 per Coding Agent task, nearly 5x. For workloads where the absolute top coding performance is mandatory, Fable 5 is still ahead. For near-frontier performance at a fraction of the cost, Grok 4.5 wins.
Versus GPT-5.5
The closest comparison. On Terminal-Bench 2.1 they are within a tenth of a point (83.3% vs 83.4%). On the Coding Agent Index they are on par. GPT-5.5 leads on the Intelligence Index overall. Grok 4.5 wins decisively on cost ($2.49 vs $5.07 per Coding Agent task) and on token efficiency (1.9M vs 6.2M tokens per task). On tau3-Banking, Grok 4.5 leads GPT-5.5 by 2 points.
Versus GLM-5.2
GLM-5.2 sits just ahead of Grok 4.5 on GDPval-AA v2 (Elo 1,513 vs 1,543). Grok 4.5 leads on the overall Intelligence Index and on tau3-Banking.
Why This Release Matters
Three things stand out from the Grok 4.5 launch that are worth tracking beyond the benchmark numbers.
xAI has reached the intelligence frontier. Grok 4.3 was not competing with Anthropic and OpenAI at the top of the capability rankings. Grok 4.5 is, with a 16-point single-generation Intelligence Index jump. That changes the competitive structure of the market from a two-player race to a three-way one.
The cost efficiency case is real, not just a pricing discount. Grok 4.5 is cheaper per token, but it is also dramatically more token-efficient per task. A model that produces the same output in 14,000 tokens when a competitor uses 67,000 is not just cheaper to run, it is faster and generates less noise in agentic workflows. Both matter at production scale.
The speed ceiling has not been hit yet. 80 tokens per second is the current figure. When xAI deploys its custom C/C++ inference stack mapped to the GB300 hardware, Musk expects that to double or more. A model already competitive on cost and quality gets significantly faster with no model update required.
Grok 4.5 is coming to Fello AI in the next few days. Once it lands, you will be able to run it alongside Claude Fable 5, Opus 4.8, GPT-5.5, Gemini, DeepSeek, and Perplexity in one native app for Mac, iPhone, and iPad, comparing the same task across every frontier model without managing separate accounts. For context on what comes after Grok 4.5, the Grok 5 article covers what xAI has signaled about the next major version.
FAQ
When was Grok 4.5 released?
xAI released Grok 4.5 on July 8, 2026. It is xAI’s first model built specifically for coding and agentic work, trained on real Cursor developer session data.
How much does Grok 4.5 cost?
API pricing is $2.00 per million input tokens and $6.00 per million output tokens. Cache hits get a 75% discount, dropping cached input to $0.50 per million. Pricing doubles for inputs longer than 200k tokens.
Is Grok 4.5 better than Claude Opus 4.8?
It depends on the task. Grok 4.5 beats Opus 4.8 on DeepSWE 1.0 and Terminal-Bench 2.1 but trails on DeepSWE 1.1 and SWE-Bench Pro, and Opus leads on the Intelligence Index. Grok 4.5 wins decisively on per-task cost and token efficiency.
What is Grok 4.5’s context window?
Grok 4.5 has a 500k-token context window, down from Grok 4.3’s 1 million tokens. It is the one clear tradeoff in an otherwise across-the-board upgrade.
Can I use Grok 4.5 in Fello AI?
Grok 4.5 is arriving in Fello AI within the next few days. Once live, you can run it next to Claude, GPT-5.5, Gemini, DeepSeek, and Perplexity in one native Mac, iPhone, and iPad app and compare the same task across every frontier model.
