Grok 4.6 pricing for categorization

Quick start

Call Grok 4.6 with the OpenAI Python SDK pointed at api.x.ai:

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.x.ai/v1",
    api_key=os.environ["XAI_API_KEY"],
)
response = client.chat.completions.create(
    model="grok-4.6",
    max_tokens=150,
    messages=[{"role": "user", "content": "Your prompt here"}]
)
# Cost: ~$0.002500 per call at standard rates
# Input:  $2.00/MTok
# Output: $6.00/MTok

How Grok 4.6 compares on cost

Per-call cost for all models scoring 50+ on categorization benchmarks. Hover for details.

Cheapest viable option: GPT 5.6 Luna at $0.000340/call (86% less, 91/100 quality).

What drives categorization cost

Classification into a fixed taxonomy is token-efficient by construction: the taxonomy is the same on every call and the output is one label. Put the taxonomy in the cached prefix and the marginal cost per item approaches the cost of the item text alone. Model tier matters less here than in any other use case tracked.

For Grok 4.6 specifically, that means output tokens cost 3.0x more than input, at $6.00 vs $2.00 per MTok. At this use case’s 800/150 token split, output accounts for 36% of the $0.002500 per-call cost.

Grok 4.6 quality benchmarks for categorization

Benchmark data for categorization is pending for Grok 4.6. Test against your own evaluation suite before moving to production.

Performance across all task types

Grok 4.6 scores by task type. Categorization highlighted.

Confidence scores are a third-party benchmark snapshot from week 29 of 2026. This snapshot is 8 weeks old, so treat the scores as directional and validate against your own evals.

About Grok 4.6

Other models from Grok

You can compare Grok 4.6 head-to-head against any other model on our model comparison pages, or track how pricing has changed over time on the pricing history tracker.

Frequently asked questions