Cheap thinking versus cheap answering: $1.10/$4.40 against $0.20/$1.20.
o4-mini · openai
gpt-5.6-luna · openai
Benchmark
| Workload Scenario | o4-mini | GPT-5.6 Luna | Price Delta |
|---|---|---|---|
| 1M input tokens (raw text) | $1.10 | $0.20 | +$0.90 |
| 1M output tokens (generation) | $4.40 | $1.20 | +$3.20 |
| 1M tokens · 70% input / 30% output mix | $2.09 | $0.50 | +$1.59 |
| Standard chat turn (4K in / 800 out, 50% cached) | $0.00627 | $0.0014 | +$0.00487 |
| Monthly scale (10K requests / day) | $1,881.00 | $420.00 | +$1,461.00 |
Negative difference = o4-mini is cheaper. Positive = GPT-5.6 Luna is cheaper.
Scaling Curve
Total cost of a token volume at 70/30 input/output split (uncached). Log-log scale.
Verdict
GPT-5.6 Luna 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 (o4-mini: 200K, GPT-5.6 Luna: 1.1M) may justify the premium for your task.
FAQ
On input, GPT-5.6 Luna is cheaper ($0.20/M vs $1.10/M). On output, GPT-5.6 Luna is cheaper ($1.20/M vs $4.40/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 $0.00627 on o4-mini and $0.0014 on GPT-5.6 Luna — GPT-5.6 Luna is 4.5× more expensive for that workload.
o4-mini supports 200,000 tokens (100K max output); GPT-5.6 Luna supports 1,050,000 (128K max output). GPT-5.6 Luna fits 5.3× more context, which matters for long documents and agents.