GPT-5.6 Terra vs GPT-5.6 Luna
OpenAI vs OpenAI · Prices checked 12 August 2026
On blended cost GPT-5.6 Luna undercuts GPT-5.6 Terra by about 10.0×: $0.500 per million tokens versus $5.00. A single 100K-input, 10K-output job costs $0.032 on GPT-5.6 Luna against $0.320 on GPT-5.6 Terra.
Of the 2 shared benchmarks, GPT-5.6 Terra takes 2. The widest gap is on LMArena Elo, where GPT-5.6 Terra leads by 14 points. Context is level at 1,050,000 tokens on each side. GPT-5.6 Terra was released Jul 2026; GPT-5.6 Luna Jul 2026.
Pricing and capacity
| Metric | GPT-5.6 Terra | GPT-5.6 Luna | Difference |
|---|---|---|---|
| Input / 1M tokens | $2.00 | $0.200 | 10.0× |
| Output / 1M tokens | $12.00 | $1.20 | 10.0× |
| Blended · 70/30calculated | $5.00 | $0.500 | 10.0× |
| Blended · 50/50calculated | $7.00 | $0.700 | 10.0× |
| Blended · 80/20calculated | $4.00 | $0.400 | 10.0× |
| Blended · 20/80calculated | $10.00 | $1.00 | 10.0× |
| Job cost · 100K + 10Kcalculated | $0.320 | $0.032 | 10.0× |
| Job cost · 1M + 100Kcalculated | $3.20 | $0.320 | 10.0× |
| Context window | 1,050,000 | 1,050,000 | level |
Input and output rates are as published by each provider. Blended and job-cost rows are calculated comparison metrics, not published prices — how they are derived.
Benchmarks
| Benchmark | GPT-5.6 Terra | GPT-5.6 Luna | Leader |
|---|---|---|---|
| GPQA Diamond | 92.9% | 92.3% | GPT-5.6 Terra |
| LMArena Elo | 1465 Elo | 1451 Elo | GPT-5.6 Terra |
Only suites both models report are listed, so a missing score is never shown as a loss. Figures are provider-published or from public leaderboards and are not independently re-run.
What each model is for
GPT-5.6 Terra
OpenAI · Balanced
The balanced everyday tier of GPT-5.6, which OpenAI describes as competitive with GPT-5.5 at a lower price. Terra is the cheaper path for scoped implementation work and first-pass review, where Sol's full reasoning budget is more than the task needs. Standard short-context pricing is $2/$12 per MTok with cached input at $0.20, roughly 60% below Sol for the same context window.
- GPT-5.5-class performance at lower cost
- Scoped implementation work
- First-pass code review
- Tool calling
GPT-5.6 Luna
OpenAI · Efficient
The fastest and most affordable GPT-5.6 tier, a deliberately low-reasoning lane for high-volume work where speed and cost are the binding constraints rather than depth. It keeps the family's 1M-token context window at $0.20/$1.20 per MTok, which is 25× cheaper than Sol on input, with cached input at $0.02.
- Lowest cost in the GPT-5.6 family, 25× cheaper input than Sol
- Fast responses
- High-volume throughput
- 1M-token context
Frequently asked questions
Which is cheaper, GPT-5.6 Terra or GPT-5.6 Luna?
On blended cost GPT-5.6 Luna undercuts GPT-5.6 Terra by about 10.0×: $0.500 per million tokens versus $5.00. A single 100K-input, 10K-output job costs $0.032 on GPT-5.6 Luna against $0.320 on GPT-5.6 Terra. Blended cost weights input at 70% and output at 30%; price your own token split for an exact answer, since the cheaper model on a blended basis is not always cheaper for your workload.
Which performs better, GPT-5.6 Terra or GPT-5.6 Luna?
Of the 2 shared benchmarks, GPT-5.6 Terra takes 2. The widest gap is on LMArena Elo, where GPT-5.6 Terra leads by 14 points. These are the figures each provider or public leaderboard publishes, not results we re-run, and a benchmark average can hide the specific behaviour your workload depends on.
Which has the larger context window, GPT-5.6 Terra or GPT-5.6 Luna?
Context is level at 1,050,000 tokens on each side. The context window caps everything in a single request: prompt, retrieved documents, conversation history and the model’s own output all share it.
Can I compare GPT-5.6 Terra and GPT-5.6 Luna on my own numbers?
Yes. Open both in the side-by-side comparison tool to see every field at once, or use the token calculator to price your actual prompt and monthly request volume against each model. Both are free and need no sign-up.
Price it for your own workload
Related matchups
Terms used on this page
Prices and scores are recomputed on every deploy from the catalogue. Prices checked 12 August 2026.