xAI · in Borg

Grok 4.5

Grok 4.5 is xAI's coding-oriented generation following the cheaper Grok 4.3.

Reasoning500,000 tokenstext, image

500,000 tokensContext window

—Median time to first token (gateway proxy)

5.9Output tokens per second (whole turn)

$2 / $6Vendor API price in / out · per 1M

Included in Borg

Download Borg Borg for Business

At a glance

Specification	Value
Provider	xAI
Family	Grok
Release date	2026-07-08 Vendor source
Context window	500,000 tokens Vendor source
Input modalities	text, image
Output modalities	text
Reasoning	Yes
Open weights	No
Borg lane ids	`grok-4.5`, `x-ai/grok-4.5`
Versions in Borg	Grok 4.3 · Grok 4.5 · Grok 4.6 · Grok 4.7
Median tokens per second (end to end) on Borg	5.9 output tokens per second

Not published by the vendor: thinking modes, knowledge cutoff, fusion roles.

What Grok 4.5 is good for

Grok 4.5 is xAI's coding-oriented generation following the cheaper Grok 4.3. Its training and launch emphasis are software engineering, agentic work and technical knowledge, with configurable reasoning rather than a non-reasoning mode. Compared with 4.6 and 4.7, it is the earlier frontier option at the same listed standard input and output prices, so it suits teams comparing coding behavior across generations. It can inspect images as well as text, making screenshots and visual bug reports practical inputs, but replies in text. Long code discussions and tool-driven engineering tasks are more natural fits than image generation or audio work. Borg includes Grok 4.5 in the subscription; select its lane when you want this generation specifically, or allow Borg Fusion to route a task to it.

agentic coding with tools
technical math and engineering questions
debugging from screenshots
long software design discussions

How Borg uses it

Grok 4.5 is available as a subscription catalog lane in Borg. The catalog does not publish which Fusion profiles route to this model, so no routing role is claimed here. You can pick it directly by setting the OpenAI-compatible gateway model field to a listed lane id. Check the current catalog for availability before building a workflow around a lane.

model: "grok-4.5"

Measured on Borg

Metric	Median	95th percentile	Samples
Time to first token (gateway proxy)	—	—	0
Tokens per second (end to end)	5.9 output tokens per second	12.7 output tokens per second	186

Output tokens per second (whole turn)

Higher is fasterFleet median 31.9 tokens/s

Grok 4.55.9 tokens/s

Time to first token (gateway proxy): time until the first streamed response event reaches the gateway; our closest available measure of time to first token
Tokens per second (end to end): tokens per second, end to end: provider-reported output tokens divided by the whole turn's elapsed time, including queueing and reasoning; not a streaming per-token rate

Completion speed covers the whole turn including queueing and reasoning time; it is not a streaming tokens-per-second figure.

Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 502 requests. Minimum 100 samples per published request metric.

Benchmarks

Benchmark	Vendor-reported score	Source
SWE Marathon resolution rate (pass@1)	29.0%	Vendor source
Terminal Bench 2.1	83.3%	Vendor source
SWE Bench Pro resolve rate	64.7%	Vendor source

Pricing

Included in Borg subscription. Vendor API list price: $2 input / $6 output per 1M tokens. Vendor price source.

Global rates for prompts below 200,000 input tokens. At 200,000 input tokens or more, all tokens use long-context rates: $4 input / $12 output / $0.6 cached input per 1M tokens. US regional inference costs 1.1× the global rate. Vendor pricing terms

Compare subscription and API pricing · Cost Savings Calculator

FAQ

What is Grok 4.5?

Grok 4.5 is listed in Borg's subscription catalog as grok-4.5. Grok 4.5 is xAI's coding-oriented generation following the cheaper Grok 4.3.

Is Grok 4.5 in Borg?

Yes. Grok 4.5 is listed in Borg's subscription catalog under the lane grok-4.5. You can choose the lane directly, or explore the available Fusion profiles. Availability and model details can change; check the current catalog before deployment.

What context window does Grok 4.5 have?

The published context window is 500,000 tokens. Consult the cited sources for details and current limits. Vendor source

Is Grok 4.5 good for coding?

Vendor-published evaluations appear in the benchmarks section; their settings may differ from your tasks. Its listed uses include agentic coding with tools, technical math and engineering questions, debugging from screenshots, long software design discussions. Try the lane for your own tasks, and compare the output with your requirements before choosing a default.

How fast is Grok 4.5 on Borg?