GPT-5.6 Luna vs Claude Sonnet 5
OpenAI's GPT-5.6 Luna and Anthropic's Claude Sonnet 5 both target production language workloads, but they price and behave differently. Here is the side-by-side.
Pricing checked against provider documentation on . How we verify
The short answer
GPT-5.6 Luna is the cheaper option — roughly 8.9x less on a blended workload, and it suits high-volume work on a tight budget. Claude Sonnet 5 justifies its premium when you need fast claude-quality reasoning at scale.
OpenAI • GPT-5.6
$0.20 / $1.20
/ 1M tokens (input / output)
After an 80% price cut in July 2026, Luna became the value outlier among frontier-family models: it outperforms the previous generation's top-end Opus tier on coding evals while costing about a fiftieth of Fable 5 per input token. This is the model to route bulk traffic through in a tiered architecture.
Anthropic • Claude 5
$2.00 / $10.00
/ 1M tokens (input / output)
The workhorse Claude tier: a full million tokens of context and adaptive thinking at $2/$10, which undercuts the previous Sonnet 4.6 generation by a third on input while raising the context ceiling fivefold. The default pick for latency-sensitive products that still want Claude's writing and tool-use behaviour.
Input price
Claude Sonnet 5 is 10.0x the price than GPT-5.6 Luna on input tokens.
Output price
Claude Sonnet 5 is 8.3x the price than GPT-5.6 Luna on output tokens — usually the side that dominates the bill.
Monthly cost at three workload sizes
Standard (non-batch, non-cached) rates. Reasoning models will exceed these figures because thinking tokens bill as output.
| Workload | GPT-5.6 Luna | Claude Sonnet 5 | Difference |
|---|---|---|---|
| Light — 1M in / 200K out | $0 | $4 | $4 |
| Moderate — 10M in / 2M out | $4 | $40 | $36 |
| Heavy — 100M in / 20M out | $44 | $400 | $356 |
Specification comparison
| Attribute | GPT-5.6 Luna | Claude Sonnet 5 |
|---|---|---|
| Input (/ 1M tokens) | $0.20 | $2.00 |
| Output (/ 1M tokens) | $1.20 | $10.00 |
| Cached input | $0.02 | $0.20 |
| Context window | 1,048,576 tokens | 1,000,000 tokens |
| Max output | 128,000 tokens | 128,000 tokens |
| Native reasoning | Yes | Yes |
| Knowledge cutoff | 2026-02 | 2026-01 |
| Relative latency | low | low |
| Open weights | No | No |
| API model ID | gpt-5.6-luna | claude-sonnet-5 |
| Status | stable | stable |
GPT-5.6 Luna: Cut 80% from the $1/$6 launch price on 2026-07-30.
Choose GPT-5.6 Luna if…
- Classification and content moderation
- High-frequency agent subtasks
- Bulk summarization
- Cost-sensitive chat features
Choose Claude Sonnet 5 if…
- Production chat and copilots
- Long-document analysis
- Tool-heavy agents needing fast turns
- Code review and PR automation
Frequently asked
Is GPT-5.6 Luna or Claude Sonnet 5 cheaper?
GPT-5.6 Luna is cheaper. On a blended 3:1 input-to-output workload it costs about 8.9x less than Claude Sonnet 5.
Which has the larger context window, GPT-5.6 Luna or Claude Sonnet 5?
GPT-5.6 Luna has the larger window at 1.05M tokens versus 1M.
Should I use GPT-5.6 Luna or Claude Sonnet 5?
Pick GPT-5.6 Luna for high-volume work on a tight budget. Pick Claude Sonnet 5 for fast claude-quality reasoning at scale. If cost dominates the decision, GPT-5.6 Luna wins; if you need the capability ceiling, benchmark both on your own evals before committing.