GPT-5.6 Luna
openai/gpt-5.6-luna
optimized for cost-sensitive workloads.
cmd --model gpt-5.6-luna
Intelligence index
52.3
Output speed
165 tok/s
Input
$0.20 /M
Output
$1.20 /M
Cache read
$0.02 /M
Agent-loop cost
$0.07 /M in
Context window
1.05M tokens
Released
July 9, 2026
Modalities
→
vs. the lineup
GPT-5.6 Luna beside its stablemates and nearest rivals. The ◆ marks the best value in each column across every row shown.
pin a rival:
| Model | Intelligence | Coding | Speed | Input $/M | Output $/M | Blended $/M | Context |
|---|---|---|---|---|---|---|---|
| GPT-5.6 Sol | 60.9◆ | 77.4◆ | 68.7◆ | $5◆ | $30◆ | $11.25◆ | 1.05M◆ |
| GPT-5.6 Terra | 56.6◆ | 76.7◆ | 117.1◆ | $2◆ | $12◆ | $4.50◆ | 1.05M◆ |
| GPT-5.5 | 56.3◆ | 74.9◆ | —◆ | $5◆ | $30◆ | $11.25◆ | 400K◆ |
| GPT-5.4 | 53.1◆ | 71.1◆ | —◆ | $2.50◆ | $15◆ | $5.63◆ | 400K◆ |
| GPT-5.6 Luna ◆ | 52.3◆ | 71.4◆ | 165◆ | $0.20◆ | $1.20◆ | $0.45◆ | 1.05M◆ |
| GPT-5.3 Codex | 45.5◆ | —◆ | 132.3◆ | $2◆ | $8◆ | $3.50◆ | 400K◆ |
| DeepSeek V4 Pro (latest)pinned | 53.2◆ | 68.8◆ | 75.3◆ | $0.66◆ | $1.98◆ | $0.99◆ | 1M◆ |
| DeepSeek V4 Flash (latest)pinned | 52◆ | 69.1◆ | 114.6◆ | $0.22◆ | $0.66◆ | $0.33◆ | 1M◆ |
coding performance
The Intelligence Index and its sub-scores, ranked against every scored model in the catalog. A metric that has not been measured for GPT-5.6 Luna has been left empty.
Coding Index
71.4
#17 of 45 scored
Terminal-Bench
80.9
#14 of 45 scored
Intelligence Index
52.3
#21 of 51 scored
Long-context reasoning
78.3
reasoning across a long context
SciCode
52.5
scientific coding
GPQA Diamond
91.1
graduate-level QA
price bands
GPT-5.6 Luna is not billed at one flat rate — the price changes with how much context a request carries. The header quotes the standard band; here is every band.
| Band | Context | Input $/M | Output $/M | Cache read $/M |
|---|---|---|---|---|
| Standard | ≤ 272K | $0.20 | $1.20 | $0.02 |
| Long context | > 272K | $0.40 | $1.80 | $0.04 |
usage calculator
How far a month of credits goes on GPT-5.6 Luna.
Input tokensfresh prompt
800Output tokensmodel reply
180Cache read tokensre-read context
50Kcost / request $0.0014 · in $0.20 · out $1.20 · cache $0.02 per M
fresh input 12%output 16%cache reads 73%
Requests / 30 days
7.3K
$10 credits ÷ $0.0014 per request
~1.5K quick fixes~291 bug fixes~48 feature PRs
what real work costs
Real coding tasks priced end to end on GPT-5.6 Luna, from a quick lookup to a full-repo agent run.
One agent task
$0.03
180K in at 75% cache hit, 12K out
What you pay
$0.14 /M
all-in across every token that task touched
Sticker input
$0.20 /M
cache reads bill at $0.02 /M instead
| Task | Tokens in · out | GPT-5.6 Luna | Muse Spark 1.2 Contributor | Claude Haiku 4.5 |
|---|---|---|---|---|
| Quick lookup / one-liner | 8K · 1K | $0.0015 | $0.0003 | $0.0065 |
| Review a 500-line PR | 60K · 4K | $0.0092 | $0.0027 | $0.04 |
| Fix a bug (agent loop) | 180K · 12K | $0.03 | $0.0072 | $0.12 |
| Refactor a module | 320K · 20K | $0.05 | $0.01 | $0.21 |
| Full-repo agent run | 900K · 45K | $0.11 | $0.03 | $0.49 |
frequently asked
How much does GPT-5.6 Luna cost?
$0.20/M input and $1.20/M output, cache reads $0.02/M. In an agent loop most input is cache-read, so the effective input rate is about $0.07/M.Which plan do I need?
Available on Go and above.
How do I switch to it?
Run
cmd --model gpt-5.6-luna, or type /model in a session and pick it. You can switch mid-session without losing context.Ship code that matches your taste
Command Code is the AI coding agent that continuously learns your taste. Start for $1.
Benchmarks from Artificial Analysis (v4.1)commandcode.ai/models/gpt-5-6-luna