Verified November 2026

Home › Compare › GPT-6 Luna vs Claude Haiku 5.5

GPT-6 Luna vs Claude Haiku 5.5

A straight cost comparison of GPT-6 Luna (OpenAI) and Claude Haiku 5.5 (Anthropic), per 1,000,000 tokens, verified November 2026.

$0.10GPT-6 Luna input / 1M
$0.10Claude Haiku 5.5 input / 1M
TieBlended price gap
EqualCheaper blended

Rate-by-rate comparison

USD per 1M tokens · November 2026
MetricGPT-6 LunaClaude Haiku 5.5Winner
Input / 1M$0.10$0.10Tie
Cached input / 1M$0.01$0.01Tie
Cache write / 1M$0.125$0.125Tie
Output / 1M$0.50$0.50Tie
Context window1,050,000200,000GPT-6 Luna
Batch APIYesYesTie
Long-context tier> 272,000> 100,000—

For per-token metrics, the lower number wins. For context window and batch support, the higher or broader option wins. A dash means the provider does not publish that rate: a missing cache-write fee means writes are not billed per token, while a missing cached-input rate means cache hits bill at the full input rate.

Blended cost: 1M input + 250K output

ModelBlended costDifference
GPT-6 Luna$0.225Tie
Claude Haiku 5.5$0.225Tie

Ten workloads, two bills

The same two models, priced across ten workloads that differ only in their input:output ratio. Nothing else changes — no retries, no system prompt, no middleware.

WorkloadGPT-6 LunaClaude Haiku 5.5Winner
Classification
100K reqs · 500 in / 10 out
$5.50$5.50Tie
Data extraction
10K reqs · 2K in / 200 out
$3$3Tie
Support chat
1K reqs · 4K in / 600 out
$0.70$0.70Tie
Code review
2K reqs · 8K in / 1.5K out
$3.10$3.10Tie
Summarisation
1K reqs · 20K in / 1K out
$2.50$2.50Tie
Agent loop
5K reqs · 12K in / 2K out
$11$11Tie
Translation
500 reqs · 15K in / 15K out
$4.50$4.50Tie
Long-document QA
200 reqs · 120K in / 800 out
$2.48$12.08GPT-6 Luna
Cached RAG
1K reqs · 50K in (45K cached) / 1K out
$1.46$1.46Tie
Nightly batch
10K reqs · 10K in / 2K out
$10$10Tie

GPT-6 Luna wins 1 of 10, Claude Haiku 5.5 wins 0 of 10. The winner is set by the token mix, not by the headline rate.

Where the two bills cross over

Both models bill identical input and output rates, so there is no crossover: the two bills are equal at every ratio, and the choice rests entirely on context window, throughput and output quality.

Cost at four scales

The same support-chat workload (4K input, 600 output per request) at four monthly volumes.

Requests / monthGPT-6 LunaClaude Haiku 5.5Difference
100$0.07$0.07Tie
1,000$0.70$0.70Tie
10,000$7$7Tie
100,000$70$70Tie

What caching does to each bill

Caching only helps the part of the prompt that repeats. This is the support-chat workload (1,000 requests, 4,000 input tokens each) with an increasing share of that prompt served from a warm cache, written once on the first request.

Cached share of the promptGPT-6 LunaClaude Haiku 5.5Winner
0%$0.70$0.70Tie
50%$0.52$0.52Tie
80%$0.412$0.412Tie
95%$0.358$0.358Tie

Both models publish a cached-read rate, so the discount is comparable: GPT-6 Luna reads cache at $0.01 per million, Claude Haiku 5.5 at $0.01.

Which one should you pick?

GPT-6 Luna wins on long-document qa, while Claude Haiku 5.5 wins on none of the ten workloads above.

Beyond price, the two differ on context window (1,050,000 vs 200,000).

Batch mode is available on GPT-6 Luna and Claude Haiku 5.5, which halves the nightly job. Both publish cached-input discounts, so a stable system prompt is worth caching on either.

FAQ

Is GPT-6 Luna cheaper than Claude Haiku 5.5?

No — they are priced identically at $0.225 on a blended 1M-input-plus-250K-output workload, with the same per-token rates. Choose on context window, throughput or output quality instead.

Which handles longer prompts, GPT-6 Luna or Claude Haiku 5.5?

Claude Haiku 5.5 has the larger context window (1,050,000 vs 200,000 tokens). Watch the long-context threshold on GPT-6 Luna and Claude Haiku 5.5: rates rise once input passes it.

Do both models support prompt caching?

GPT-6 Luna bills cached reads at $0.01 per 1M; Claude Haiku 5.5 at $0.01.

Related comparisons

Run your own token volumes