About
GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows, providing capable reasoning for...
Related Models
Pricing
Input
$1.00
per 1M tokens
Output
$6.00
per 1M tokens
Blended
$2.25
per 1M tokens
Cheaper than 26% of models. Median price is $0.66/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$2.25
Monthly
$67.50
vs. Similar Models
Z.ai: GLM 5.2Q:-0.1
$0.65-71%
OpenAI: GPT-5.4Q:+0.2
$5.63+150%
GPT-5.6 Terra (xhigh)Q:+0.4
$5.63+150%
Muse Spark 1.1 (xhigh)Q:-0.6
$2.00-11%
Performance
241
tokens/sec
Faster than 93% of models
62.14
seconds
Faster than 3% of models
62.14
seconds
Faster than 5% of models
Market Median
99 tok/s
143% faster
Median TTFT
1.11s
5513% slower
Throughput/Dollar
107
tok/s per $/1M
Speed Comparison
Google: Gemini 3.5 Flash
241 tok/s+0%
OpenAI: o3 Mini High
235 tok/s-2%
Qwen3.5 Omni Flash
250 tok/s+4%
Context Window
1.1M
tokens
Larger than 96% of models
Max Output
128K
tokens
12% of context
Benchmarks
MMLU-ProNot evaluated
GPQA Diamond
91.1%
HLE
37.2%
LiveCodeBenchNot evaluated
SciCode
52.5%
TerminalBench HardNot evaluated
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBenchNot evaluated
Long Context Recall
74.0%
Tau2Not evaluated
Market AverageTop Score