Claude Haiku 5.5 vs GPT-6 Luna pricing
Side-by-side LLM API pricing.Input rates are equal ($0.10 vs $0.10 /1M), and output rates are equal ($0.50 vs $0.50 /1M).
| Attribute | Claude Haiku 5.5 | GPT-6 Luna |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Tier | Budget | Budget |
| Status | GA | GA |
| Context window | 1M | 1.05M |
| Input /1M | $0.10 | $0.10 |
| Output /1M | $0.50 | $0.50 |
| Cached input /1M | $0.01 | $0.01 |
| Batch input /1M | $0.05 | $0.05 |
| Batch output /1M | $0.25 | $0.25 |
| Blended (avg in+out) | $0.30 | $0.30 |
| 1M in + 1M out | $0.6000 | $0.6000 |
| 1M in (90% cached) + 100K out | $0.0690 | $0.0690 |
List prices, USD, directional. Rates are provider list prices per 1M tokens and are meant for comparison, not billing. Batch rates shown as 50% off are derived where a provider offers batch but does not publish a separate figure. Preview, promo, intro, peak/off-peak, long-context, and third-party-host prices are labeled where they apply. Token counts vary by tokenizer, so per-token price is not always a like-for-like cost. Always confirm with the provider before relying on a number. Prices as of 2026-11-02.
Tokenizer caveat: Claude Haiku 5.5 uses a tokenizer that produces more tokens per unit of text, so identical content is billed as a different number of tokens on each model. Per-token price is therefore not a like-for-like cost — compare cost per task, not just the sticker rate.
Which should you pick?
For a balanced job (1M input + 1M output), both models cost $0.6000. If your prompts are large and reused, prompt caching changes the maths — the cache-heavy row above shows a 90%-cached input scenario. If your work can run asynchronously, both providers' batch rates cut the bill further where offered.
Full price matrices: Claude Haiku 5.5 pricing → · GPT-6 Luna pricing → · Back to the LLM pricing hub →