Model comparison

Claude Haiku 5.5 vs GPT-5.6 Luna

Anthropic's Claude Haiku 5.5 against OpenAI's GPT-5.6 Luna: vendor list prices, limits, abilities and the Artificial Analysis indexes side by side - and what OpenSmartRoute does when both are available.

CheaperClaude Haiku 5.5

Claude Haiku 5.5 lists at $0.2 per million tokens blended against $0.45 for GPT-5.6 Luna - 2.3x less.

Stronger on intelligenceGPT-5.6 Luna

GPT-5.6 Luna scores 37 on the intelligence index against 36 - 1 points apart.

Longer contextGPT-5.6 Luna

GPT-5.6 Luna reads 1.1M tokens at once, Claude Haiku 5.5 1M.

NewerClaude Haiku 5.5

Claude Haiku 5.5 was released about 3 months later.

Side by side

Vendor list prices per million tokens and published limits; indexes are Artificial Analysis composites (0-100).

Specification	Claude Haiku 5.5	GPT-5.6 Luna
Input price / 1M tokens	$0.10	$0.20
Output price / 1M tokens	$0.50	$1.20
Blended (3 in : 1 out)	$0.20	$0.45
Cached input / 1M	$0.01	$0.02
Context window	1M	1.1M
Max output	128K	128K
Input	text, image, file	file, image, text
Output	text	text
Tool calls
Structured output
Reasoning
Open weights
Intelligence index	36	37
Coding index	-	71
Agentic index	-	42
Released	Oct 7, 2026	Jul 9, 2026
Available on this deployment

What a month costs

List price times tokens for three typical workloads; your mix of input and output decides which model's pricing suits you.

Workload	Claude Haiku 5.5	GPT-5.6 Luna	Difference
Support chat20M tokens in, 5M out a month - a help desk answering a few thousand conversations.	$4.50	$10.00	$5.50 less on Claude Haiku 5.5
Document summaries100M in, 5M out - long inputs, short answers.	$12.50	$26.00	$13.50 less on Claude Haiku 5.5
Coding assistant60M in, 20M out - a team's editor assistant, output-heavy.	$16.00	$36.00	$20.00 less on Claude Haiku 5.5

What the router does with the pair

Claude Haiku 5.5 takes the short and general requests - at $0.2 blended it is the default - and GPT-5.6 Luna the ones whose complexity or quality floor rules the cheaper model out. The split is per request, not per app, and every decision carries the scores for both.

Price a request Playground

yaml

# targets.yaml - both models as routing targets (metadata.model links each to the catalogue entry)
targets:
  - { id: claude-haiku-5.5, kind: llm, metadata: { model: anthropic/claude-haiku-5.5 } }
  - { id: gpt-5.6-luna, kind: llm, metadata: { model: openai/gpt-5.6-luna } }

# rules.yaml - the cheaper model by default, the stronger one when the request is hard
rules:
  - name: simple-requests-stay-cheap
    when: { max_complexity: 0.45 }
    prefer: [claude-haiku-5.5]
    weight: 0.8
  - name: hard-requests-escalate
    when: { min_complexity: 0.45 }
    prefer: [gpt-5.6-luna]
    weight: 0.8

Frequently asked

Is Claude Haiku 5.5 cheaper than GPT-5.6 Luna?: Yes. Claude Haiku 5.5 lists at $0.10 in / $0.50 out per million tokens, GPT-5.6 Luna at $0.20 in / $1.20 out; blended, Claude Haiku 5.5 is the cheaper one.
Which is better, Claude Haiku 5.5 or GPT-5.6 Luna?: On the Artificial Analysis intelligence index GPT-5.6 Luna scores higher (37 against 36). "Better" depends on the task: the coding and agentic indexes below and the abilities row may point the other way for your requests.
Which has the longer context window?: GPT-5.6 Luna - 1.1M tokens against 1M.
Can I use Claude Haiku 5.5 and GPT-5.6 Luna together?: Yes. Declare both as targets and OpenSmartRoute picks one per request on quality for the task, price and your rules - the cheaper model for the requests it can answer, the stronger one for the rest - and records why in the trace.

Also compared

Every Anthropic and OpenAI model, with prices, is on the vendor pages; the rankings show what the router actually picks.

For the business

What it means for which models you pay for.

Catalogue of every vendor
The right model, not the famous one
Price, context, abilities and benchmark indexes side by side, with the model that fits each task named.
The rankings
89% less
Lower cost per request
On the 150-prompt benchmark the router spent 89% less than sending every prompt to the reference model, and the judged quality went up, not down.
Read the benchmark
Per million tokens, every model
Prices you can compare
The catalogue prints input and output prices and what three typical monthly workloads cost on each.
The model catalogue

Which LLM is the cheapest for our workload?: The best-for pages rank the live catalogue per task by the index that measures it, and the router picks the cheapest model that meets your quality rule for each request.
How much can we save on our LLM bill?: On the public benchmark the router saved 89% against sending everything to the reference model. Import last month's provider bill on the bill page and the estimate is computed on your own traffic before you change a line of code.
How often do model prices change?: The catalogue is refreshed from the vendors' price lists; each model page shows the date of its data.

Stop choosing between Claude Haiku 5.5 and GPT-5.6 Luna. Route to both.

Declare the pair as targets and OpenSmartRoute picks per request - the cheaper one where it holds up, the stronger one where it matters - with the comparison in every trace.

Create a free account Try a request

Free plan, no card. Fifteen thousand decisions a month with the full trace.

Specification

Claude Haiku 5.5

GPT-5.6 Luna

Input price / 1M tokens

$0.10

$0.20

Output price / 1M tokens

$0.50

$1.20

Blended (3 in : 1 out)

$0.20

$0.45

Cached input / 1M

$0.01

$0.02

Context window

1.1M

Max output

128K

Input

text, image, file

file, image, text

Output

text

Tool calls

Structured output

Reasoning

Open weights

Intelligence index

Coding index

Agentic index

Released

Oct 7, 2026

Jul 9, 2026

Available on this deployment

Workload

Claude Haiku 5.5

GPT-5.6 Luna

Difference

Support chat20M tokens in, 5M out a month - a help desk answering a few thousand conversations.

$4.50

$10.00

$5.50 less on Claude Haiku 5.5

Document summaries100M in, 5M out - long inputs, short answers.

$12.50

$26.00

$13.50 less on Claude Haiku 5.5

Coding assistant60M in, 20M out - a team's editor assistant, output-heavy.

$16.00

$36.00

$20.00 less on Claude Haiku 5.5

# targets.yaml - both models as routing targets (metadata.model links each to the catalogue entry) targets: - { id: claude-haiku-5.5, kind: llm, metadata: { model: anthropic/claude-haiku-5.5 } } - { id: gpt-5.6-luna, kind: llm, metadata: { model: openai/gpt-5.6-luna } } # rules.yaml - the cheaper model by default, the stronger one when the request is hard rules: - name: simple-requests-stay-cheap when: { max_complexity: 0.45 } prefer: [claude-haiku-5.5] weight: 0.8 - name: hard-requests-escalate when: { min_complexity: 0.45 } prefer: [gpt-5.6-luna] weight: 0.8

Frequently asked

Is Claude Haiku 5.5 cheaper than GPT-5.6 Luna?

Yes. Claude Haiku 5.5 lists at $0.10 in / $0.50 out per million tokens, GPT-5.6 Luna at $0.20 in / $1.20 out; blended, Claude Haiku 5.5 is the cheaper one.

Which is better, Claude Haiku 5.5 or GPT-5.6 Luna?

On the Artificial Analysis intelligence index GPT-5.6 Luna scores higher (37 against 36). "Better" depends on the task: the coding and agentic indexes below and the abilities row may point the other way for your requests.

Which has the longer context window?

GPT-5.6 Luna - 1.1M tokens against 1M.

Can I use Claude Haiku 5.5 and GPT-5.6 Luna together?

Yes. Declare both as targets and OpenSmartRoute picks one per request on quality for the task, price and your rules - the cheaper model for the requests it can answer, the stronger one for the rest - and records why in the trace.

Claude Haiku 5.5 vs GPT-5.6 Luna

Side by side

What a month costs

What the router does with the pair

Frequently asked

Also compared

What it means for which models you pay for.

The right model, not the famous one

Lower cost per request

Prices you can compare

Stop choosing between Claude Haiku 5.5 and GPT-5.6 Luna. Route to both.

Claude Haiku 5.5 vs GPT-5.6 Luna

Side by side

What a month costs

What the router does with the pair

Frequently asked

Also compared

What it means for which models you pay for.

The right model, not the famous one

Lower cost per request

Prices you can compare

Stop choosing between Claude Haiku 5.5 and GPT-5.6 Luna. Route to both.