Claude Haiku 5.5 vs GPT-6 Luna
Official specifications and direct API prices. Your task decides the right fit.
Claude Haiku 5.5
Focused extraction, classification and routing tasks.
GPT-6 Luna
Compact text tasks and high-volume applications.
Key differences
Everyday overview| Specification | MODEL AClaude Haiku 5.5 | MODEL BGPT-6 Luna |
|---|---|---|
| Shared context window | 1,000,000 | 1,050,000 |
| Input token limit | Not verified | 922,000 |
| Maximum output tokens | 128,000 | 128,000 |
| Accepted input | text, image | text, image |
| Output | text | text |
| Reasoning | Supported | Supported |
Unknown fields are not evidence of equal capabilities. Specifications describe support, not comparative answer quality.
No reliable data available
Comparable results with confirmed reuse permission have not been published here yet.
A published result must include the exact model version, benchmark version, original value and unit, test date, conditions and source. No scores or overall rankings are inferred.
Official API pricing
USD / 1M text tokens · Standard| Token rate / condition | MODEL AClaude Haiku 5.5 | MODEL BGPT-6 Luna |
|---|---|---|
| Base input / 1M | $0.10 | $0.10 |
| Base output / 1M | $0.50 | $0.50 |
| Cached input read / 1M | $0.01 | $0.01 |
| Higher tier above input tokens | 100,000 | 272,000 |
| Higher-tier input / 1M | $0.50 | $0.20 |
| Higher-tier output / 1M | $2.50 | $0.75 |
| Higher-tier cache read / 1M | $0.05 | $0.02 |
| Price review required after | No expiry documented | No expiry documented |
Official pricing source · verified 2026-10-09
Text tokens only. Output includes thinking tokens. Cache reads are a subset of input. Cache creation, tools and media fees are excluded.
Official pricing source · verified 2026-10-09
Text tokens only. Billable output includes reasoning tokens where applicable. Excludes cache creation, tools, media and regional or processing-tier uplifts.
Default 30-minute cache writes use 1.25× uncached input pricing as a replacement rate. V1 live trials request explicit-only caching with no breakpoints.
A small prompt. A clearer budget.
Same token counts for both models. Standard paid text rates in USD. Output must include all billable reasoning tokens.
Cached input is included in total input. No cache creation/storage, tools, media, taxes or regional/priority/Batch fees. Tokenizers differ; equal token counts do not imply equal text length.
Base tier · Official rates · verified 2026-10-09
Base tier · Official rates · verified 2026-10-09
See how your prompt behaves.
Choose an enabled official model pair in the Playground. Your maximum reservation is shown before running; confirmed usage determines the final charge.