GPT-5.6 Terra Max vs Grok 4.6 Xhigh: speed, latency and cost

There is very little daylight between GPT-5.6 Terra (Max) (OpenAI) and Grok 4.6 (Xhigh) (SpaceXAI) on the overall index. GPT-5.6 Terra Max streams output at about 116 tokens per second, 1.9 times the rate of Grok 4.6 Xhigh. No hands-on testing went into this write-up; it summarises the recorded benchmark, price and speed data for both models.

GPT-5.6 Terra Max is the faster writer

GPT-5.6 Terra Max generates about 116 tokens per second and Grok 4.6 Xhigh about 62 tokens per second. GPT-5.6 Terra Max is ahead by 87% on throughput. Before any text appears, GPT-5.6 Terra Max takes 3.4 minutes and Grok 4.6 Xhigh 47 seconds. On total time, Grok 4.6 Xhigh wins at ordinary lengths; GPT-5.6 Terra Max's higher throughput only pays off beyond about 21,000 tokens. A first-token wait of that length suits background work better than live conversation.

Add the two numbers up for an 800-token reply and Grok 4.6 Xhigh finishes first: 3.5 minutes for GPT-5.6 Terra Max, 60 seconds for Grok 4.6 Xhigh.

How the intelligence index compares

GPT-5.6 Terra Max records 32.3 on the intelligence index and Grok 4.6 Xhigh records 31.0. That is a modest edge: real, but small enough that the right set of tasks could reverse it.

Across the 109 active models with an intelligence score in the catalogue, GPT-5.6 Terra Max ranks 46th and Grok 4.6 Xhigh ranks 49th.

For coding or maths work specifically, the catalogue has no dedicated index on either side yet.

Release timing

Grok 4.6 Xhigh arrived on August 12, 2026, 34 days after GPT-5.6 Terra Max (July 9, 2026). That is close enough to count them as the same generation.

Running costs

GPT-5.6 Terra Max lists $2.00 per million input tokens and $12.00 per million output tokens; Grok 4.6 Xhigh lists $2.00 and $6.00. GPT-5.6 Terra Max costs twice as much per output token.

Take 1 million input tokens and 200,000 output tokens: that comes to $4.40 on GPT-5.6 Terra Max and $3.20 on Grok 4.6 Xhigh. Scale it up to a chat app with 50,000 sessions of 800 input and 300 output tokens and the bill is $260 for GPT-5.6 Terra Max against $170 for Grok 4.6 Xhigh. On a 3:1 blend of input to output tokens, that works out to $4.50 versus $3.00 per million.

Value for money

A blunt value measure, blended price over intelligence score, puts GPT-5.6 Terra Max at $0.14 and Grok 4.6 Xhigh at $0.097 per point. Grok 4.6 Xhigh delivers each index point for 31% less; whether GPT-5.6 Terra Max's extra 1.3 points are worth the premium depends on how often your tasks need them.