Grok 4.6 pricing for chain of thought

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=5000,
    messages=[{"role": "user", "content": "Your prompt here"}]
)
# Cost: ~$0.0360 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 chain of thought benchmarks. Hover for details.

Cheapest viable option: Claude Sonnet 5 at $0.0560/call (-56% less, 90/100 quality).

What drives chain of thought cost

Reasoning tasks invert the usual ratio: the model produces far more tokens than it consumes, because the intermediate steps are the work. Output pricing dominates completely, and models that reason at length cost more than their headline rate suggests. Comparing reasoning models on input price is actively misleading; compare them on output.

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 3,000/5,000 token split, output accounts for 83% of the $0.0360 per-call cost.

Grok 4.6 quality benchmarks for chain of thought

Benchmark data for chain of thought 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. Chain of Thought 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