The $1.25/$10 2025 anchor versus today's $2/$12 default.
gpt-5.1 · openai
gpt-5.6-terra · openai
Benchmark
| Workload Scenario | GPT-5.1 | GPT-5.6 Terra | Price Delta |
|---|---|---|---|
| 1M input tokens (raw text) | $1.25 | $2.00 | −$0.75 |
| 1M output tokens (generation) | $10.00 | $12.00 | −$2.00 |
| 1M tokens · 70% input / 30% output mix | $3.875 | $5.00 | −$1.125 |
| Standard chat turn (4K in / 800 out, 50% cached) | $0.0108 | $0.014 | −$0.00325 |
| Monthly scale (10K requests / day) | $3,225.00 | $4,200.00 | −$975.00 |
Negative difference = GPT-5.1 is cheaper. Positive = GPT-5.6 Terra is cheaper.
Scaling Curve
Total cost of a token volume at 70/30 input/output split (uncached). Log-log scale.
Verdict
GPT-5.1 is cheaper on both input and output rates, so it costs less at every input/output mix. Price alone still isn't the whole decision: capability, latency and context limits (GPT-5.1: 400K, GPT-5.6 Terra: 1.1M) may justify the premium for your task.
FAQ
On input, GPT-5.1 is cheaper ($1.25/M vs $2.00/M). On output, GPT-5.1 is cheaper ($10.00/M vs $12.00/M). The same model is cheaper on both sides, so it wins at every mix.
A chat-style request (4,000 input + 800 output tokens, 50% cached) costs $0.0108 on GPT-5.1 and $0.014 on GPT-5.6 Terra — GPT-5.1 is 1.3× cheaper for that workload.
GPT-5.1 supports 400,000 tokens (128K max output); GPT-5.6 Terra supports 1,050,000 (128K max output). GPT-5.6 Terra fits 2.6× more context, which matters for long documents and agents.