Claude Haiku 5.5 vs GPT-6 Luna
Claude Haiku 5.5 and GPT-6 Luna are the two cheapest big-lab models, both listed at $0.10 in and $0.50 out per 1M tokens. The price is identical; the bill is not.
Our verdict
With identical list prices, judge by cost per finished job. Claude Haiku 5.5 scores higher on the Artificial Analysis index (43 against 37, both at max effort), but Artificial Analysis found it writes about three times as many output tokens per task (about 162K against about 50K per index task), so at the same per-token price a finished job costs roughly three times more. Pick GPT-6 Luna for high-volume work where the bill matters, and Claude Haiku 5.5 when quality on harder jobs matters more. Keep Claude Haiku 5.5 prompts under 100K tokens, because above that its rate becomes $0.50 / $2.50; GPT-6 Luna adds its surcharge above 272K tokens.
Benchmarks measure test tasks, not your business. Run a real job through both models before you commit.
Side by side
| Claude Haiku 5.5 | GPT-6 Luna | |
|---|---|---|
| Lab | Anthropic | OpenAI |
| List price per 1M tokens (input / output) | $0.10 / $0.50 | $0.10 / $0.50 |
| Scheduled price | - | - |
| Price notes | For prompts up to 100K tokens; over 100K the rate is $0.50 input and $2.50 output. | Cached input $0.01 per 1M; prompts over 272K input tokens bill the whole request at $0.20 / $0.75. |
| Example job at list price | $0.20 | $0.20 |
| Context window | 1,000,000 tokens | 1,050,000 tokens |
| Released | ||
| Open weights | No | No |
| Artificial Analysis Intelligence Index | 43 (Oct 8, 2026 capture)at max effort | 37 (Sep 23, 2026 capture)at max effort |
1M tokens in and 200K tokens out. List price only: models write different amounts, so a finished job can cost more or less. (1,000,000 / 200,000)
Common questions
Which is cheaper, Claude Haiku 5.5 or GPT-6 Luna?
They list at the same $0.10 in and $0.50 out per 1M tokens, but Artificial Analysis found Haiku 5.5 uses about three times as many output tokens per task at max effort, so a finished job on Luna usually costs less.
Is Claude Haiku 5.5 smarter than GPT-6 Luna?
On the independent Artificial Analysis index Haiku 5.5 scores 43 and Luna 37, both at max effort, so Haiku is ahead on quality. It pays for that with longer outputs.
What are the price catches?
Haiku 5.5 charges $0.50 in and $2.50 out per 1M tokens for prompts over 100K tokens. Luna bills the whole request at $0.20 in and $0.75 out for prompts over 272K tokens.
Sources
More model comparisons
All model comparisonsSee every model on the AI model tracker
Not sure which model your business needs?
Dapols picks the models and tools for your budget and shows the bill before you spend.