Claude Haiku 5.5 vs GPT-6 Luna: API Pricing Compared
Official prices, verified 2026-10-24. Every conclusion on this page is computed from the price data — if the prices change, the verdict changes.
| Claude Haiku 5.5 (Anthropic) | GPT-6 Luna (OpenAI) | |
|---|---|---|
| Input / 1M | $0.1 | $0.1 |
| Output / 1M | $0.5 | $0.5 |
| Cached input / 1M | $0.01 | $0.01 |
| Batch tier | $0.05 / $0.25 | Not published |
| Context window | 1,048,576 | 1,050,000 |
| Price verified | 2026-10-24 | 2026-10-24 |
Monthly cost at three workloads
| Workload | Claude Haiku 5.5 | GPT-6 Luna |
|---|---|---|
| Chat assistant 2,000 input / 500 output tokens per call, 10,000 calls a month | $4.50 | $4.50 |
| Document processing 50,000 input / 2,000 output tokens per call, 1,000 calls a month | $6.00 | $6.00 |
| Agent loop with prompt caching 10,000 fresh input / 100,000 cached input / 2,000 output tokens per call, 500 calls a month | $1.50 | $1.50 |
The verdict, from the data
- Input is priced identically at $0.1 per 1M tokens.
- Output is priced identically at $0.5 per 1M tokens.
- At a blended workload of 1M input + 250K output tokens the two models cost exactly the same ($0.225) — the decision comes down to caching, batch and context, below.
- Only Claude Haiku 5.5 publishes a batch tier ($0.05 / $0.25), which halves its rates for latency-tolerant work.
Run your own numbers in the homepage calculator, or open the dedicated calculators for Claude Haiku 5.5 and GPT-6 Luna.