xAI

Grok 3 Mini

Cost-efficient Grok variant with good general-purpose quality

Model Summary

Family

Grok

Version

3.0

Parameters

190B (est.)

Parameter counts for closed models are estimates; vendors rarely publish exact sizes.

VRAM Requirements by Quantization

Memory needed to serve Grok 3 Mini for inference, including a 20% overhead for activations and KV cache.

PrecisionVRAM neededSmallest single GPU that fits
INT4 (4-bit)106.17 GBAMD Instinct MI250X (128 GB)
INT8 (8-bit)212.34 GBAMD Instinct MI325X (256 GB)
FP16 (16-bit)424.68 GBMulti-GPU required
FP32 (32-bit)849.37 GBMulti-GPU required

Recommended GPU Configurations

Cheapest on-demand configurations to serve Grok 3 Mini at 8-bit (212 GB VRAM).

2x AMD Instinct MI250X

256 GB total VRAM · CDNA 2

~$5.00/h

14x NVIDIA T4

224 GB total VRAM · Turing · multi-node

~$7.00/h

1x AMD Instinct MI325X

256 GB total VRAM · CDNA 3

~$8.00/h

Quick GPU Planning

Use the calculator pre-filled with this exact version to estimate memory, speed, and compute requirements in a few clicks.

Access Pre-filled Calculator

Frequently Asked Questions

How much VRAM do you need to run Grok 3 Mini?

With an estimated 190B parameters, Grok 3 Mini needs roughly 212 GB of VRAM in 8-bit (INT8), 106 GB in 4-bit, and 425 GB in FP16, including a 20% overhead for activations and KV cache.

Which GPUs can run Grok 3 Mini?

At 8-bit quantization, the most cost-effective option is 2x AMD Instinct MI250X (256 GB combined VRAM, around $5.00/hour on-demand). Higher-end cards like the NVIDIA B200 or AMD MI355X reduce the GPU count needed.

Can Grok 3 Mini run on a single GPU?

Yes. In 8-bit, a single AMD Instinct MI325X (256 GB) fits the model.

How much does it cost to serve Grok 3 Mini in the cloud?

Renting 2x Instinct MI250X costs on the order of $5.00/hour, i.e. about $3,650/month running 24/7. Actual prices vary by provider and commitment; spot and reserved capacity can be significantly cheaper.

Other Grok Versions