RentGPU RentGPU
Encontrar GPU →
by xAI

Grok 3.

multimodal closed 1M ctx Transformer Quality 84.0
Cheapest input
$3.0/M
on xAI API
Cheapest output
$15.0/M
on xAI API
Fastest
112 tok/s
on OpenRouter
Hosted equiv.
~$5.4/hr
@ 100 tok/s on xAI API

xAI's frontier model with built-in DeepSearch + real-time X integration.

Hosted API only

No self-host path — closed weights.

Grok 3's weights aren't published. Use it via the access providers below.

Where to use it

Cheapest hosted endpoints.

Provider Access $/M in $/M out
xAI API api direct $3.0 $15.0 Launch ↗
Capability snapshot Full benchmarks →

What it's best at.

Math 93.3
Coding 88.4
Graduate-level science 84.6
Reasoning 79.9
Performance

Speed across providers.

Tokens/sec and time-to-first-token measured against the same prompt template on each provider's API.

Provider Tokens/sec TTFT Total
OpenRouter 111.5 5086 ms 6501 ms
Sources

Official references.

Launch post ↗ Demo / Chat ↗ Docs ↗
Best for

Workloads.

Run LLMs (inference / serving)
FAQ

Frequently asked.

How do I run Grok 3?
Grok 3 is a closed-source API model. The cheapest way to access it is through the API providers listed on this page (direct API, aggregators, and hosted chat UIs).
Where can I access Grok 3?
Grok 3 is available via xAI API, X (Premium+), OpenRouter. Each access option lists its own pricing (per million tokens or hourly hosting).
How much does it cost to run Grok 3?
API pricing starts at $3.0/M input tokens and $15.0/M output tokens. Self-hosting cost depends on the GPU you rent — see the Run It Yourself tab.
Is Grok 3 open-source or proprietary?
Grok 3 is a proprietary model from xAI. Access is API-only — there are no public weights to download.