By the toolist.ai editors. Researched from public sources, not hands-on tested by toolist.ai. Updated November 3, 2026; prices checked November 3, 2026.

GPT-6 Luna vs GPT-6.1 Terra

Verdict

Use GPT-6 Luna for simple, high-volume jobs and GPT-6.1 Terra for anything that needs real reasoning. Luna costs a tenth of Terra's token prices ($0.10 and $0.50 per million against $1 and $5). Terra scores higher on all five Artificial Analysis rows both have, by the widest margin on Terminal-Bench-Science (58% against 9%), and at medium effort, the API default on both, it scores 38 against 18 on the Intelligence Index.

Verdict based on published prices, limits and benchmarks, not on our own tests.

Choose GPT-6 Luna if

  • Cost matters most: $0.10 and $0.50 per million tokens against $1 and $5
  • You send many repeated prompts: cached input costs $0.01 per million against $0.05
  • You want typed answers from OpenAI's Decisions API, where GPT-6 Luna is the only supported model
  • You use Codex or ChatGPT Work on the Free or Go plan, where GPT-6 Luna is the model offered; GPT-6.1 Terra needs Plus or above

Choose GPT-6.1 Terra if

  • You run agents or terminal work: 56% against 13% on Terminal-Bench 4.0
  • You need harder reasoning: 51 against 28 on the Intelligence Index at max effort and 47% against 22% on HLE-Diamond
  • You want the later knowledge cutoff (June 1, 2026 against May 18, 2026)

Price

Input costs 10 times as much on GPT-6.1 Terra as on GPT-6 Luna. Output costs 10 times as much on GPT-6.1 Terra as on GPT-6 Luna.

Per million tokensGPT-6 LunaGPT-6.1 Terra
Input$0.10 source$1 source
Cached input$0.01 source$0.05 source
Output$0.50 source$5 source

GPT-6 Luna: Prompts over 272,000 input tokens are billed at twice the input and cache rates and 1.5 times the output rate for the whole request. Cache writes cost $0.125 per million tokens (30-minute cache) or $0.25 (12-hour cache). Batch and Flex cost half; Fast mode costs double. OpenAI lists no Ultrafast price for it.

GPT-6.1 Terra: Prompts over 272,000 input tokens cost twice the input and cache rates and 1.5 times the output rate for the whole request. Batch and Flex cost half; Fast mode costs double. Cache writes cost $1.25 per million tokens with 30-minute retention and $2.50 with 12-hour retention.

Worked example: what a month costs

We priced three monthly workloads at the standard rates. The provider does not publish these figures.

WorkloadGPT-6 LunaGPT-6.1 Terra
Support chatbot
100,000 replies, each 3,000 input and 400 output tokens, no caching (300M in, 40M out)		 $50.00$500
Document Q&A
5,000 questions over the same 60,000-token document set (90% cache hits), 800 output tokens each (300M in, 4M out)		 $7.70$63.50
Coding agent
200 tasks, each 1.5 million input tokens (90% cache hits) and 50,000 output tokens (300M in, 10M out)		 $10.70$93.50

Price your own workload

ModelCost a month
GPT-6 Luna $7.50
GPT-6.1 Terra $75.00

Standard API prices, no batch or priority discounts. Cached input is billed at the cache-read price; cache-write charges are not included. Reasoning models bill their thinking as output tokens.

Limits and release dates

GPT-6 LunaGPT-6.1 Terra
ReleasedSeptember 22, 2026 sourceOctober 26, 2026 source
Context window1,050,000 tokens source1,050,000 tokens source
Maximum output128,000 tokens source128,000 tokens source
Knowledge cutoffMay 18, 2026 sourceJune 1, 2026 source
Input and outputText and images in; text out sourceText and images in; text out source
API model namegpt-6-lunagpt-6.1-terra

Benchmarks

Scores as published by the provider or measured by Artificial Analysis; higher is better. A dash means no score was published.

BenchmarkGPT-6 LunaGPT-6.1 Terra
Artificial Analysis Intelligence Index
max effort		 28.0 source51.0 source
Intelligence Index at everyday effort
medium effort, the API default		 18.0 source38.0 source
Terminal-Bench 4.0
max effort		 13.0% source56.0% source
HLE-Diamond
max effort		 22.0% source47.0% source
Terminal-Bench-Science 0.1
max effort		 9.0% source58.0% source
Output speed 111.1 tokens/s (max effort, Artificial Analysis; no medium-effort figure is published) source88.7 tokens/s (medium effort, Artificial Analysis) source

Benchmark and speed data from Artificial Analysis where linked.

Read more

Sources