Skip to main content
BenchLM
Data

GPT-6 Luna vs MAI Code 2

Updated November 17, 2026. We do not rank this pair: at least one has no public score. Public scores include evidence status and uncertainty.

Share or export
Share on XLinkedInAPI/MCP

Decision reading

At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. 1 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Model A
OpenAI logo

OpenAI

61.01/100

Supported · Public rank #50

90% interval 49.4–72.6

Model B
Microsoft logo

Microsoft

—

Evidence status unavailable

90% interval unavailable

API price per 1M tokens

GPT-6 Luna: $0.10 input / $0.50 output

MAI Code 2: $1 input / $5 output

Context window

GPT-6 Luna: 1.05M

MAI Code 2: 1.05M

Released

GPT-6 Luna: September 22, 2026

MAI Code 2: November 17, 2026

Shared results
1
GPT-6 Luna only
28
MAI Code 2 only
2
Like-for-like categories
0 / 8
Supported: GPT-6 LunaHow the comparison works

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

  • Chat turn cost

    1K fresh input + 500 output tokens

    GPT-6 Luna

    GPT-6 Luna has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
  • Cache-heavy agent loop cost

    200K cached + 20K fresh input + 10K output tokens

    GPT-6 Luna

    GPT-6 Luna has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
Show secondary and unsupported calls
  • Repository review cost

    50K fresh input + 3K output tokens

    GPT-6 Luna

    GPT-6 Luna has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    MAI Code 2 is not ranked on the public lane for coding, so no winner is named for coding.

    Confidence: limited
  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    MAI Code 2 is not ranked on the public lane for agentic, so no winner is named for agentic.

    Confidence: limited
  • Long documents

    Prompts that approach the documented context limit

    No clear pick

    The documented context windows are equal.

    Confidence: documented

Which one for a specific job

Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.

The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.

52.2GPT-6 Luna—MAI Code 2

Not comparable · BenchAlign v5.8

The coding row is not comparable on the public lane: at least one model is not measured or not ranked there.

Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.

Same basis rules as the category table below

Or describe your own job

The LLM Selector reads what you need a model for and returns a shortlist, ranked in the same order the job pages publish.

Describe the task and constraints. This selector recommends models; it cannot carry out the task or access your files and accounts.

We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

A shared-evidence shape is not available.

BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign v5.8 lane for agentic, coding, reasoning, multimodal, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.

Agentic

Not comparable
GPT-6 Luna
54.8
Supported · #46/140
MAI Code 2
Not ranked
Basis
BenchAlign v5.8 lane · 8 vs 1 public rows
Reading
Not comparable

Coding

Not comparable
GPT-6 Luna
52.2
Supported · #38/163
MAI Code 2
Not ranked
Basis
BenchAlign v5.8 lane · 4 vs 2 public rows
Reading
Not comparable

Reasoning

Not comparable
GPT-6 Luna
54.3
Supported · #46/161
MAI Code 2
Not ranked
Basis
BenchAlign v5.8 lane · 5 vs 0 public rows
Reading
Not comparable

Multimodal

Not comparable
GPT-6 Luna
62.6
Supported · #26/67
MAI Code 2
Not ranked
Basis
BenchAlign v5.8 lane · 5 vs 0 public rows
Reading
Not comparable

Knowledge

Not comparable
GPT-6 Luna
62.4
Supported · #40/181
MAI Code 2
Not ranked
Basis
BenchAlign v5.8 lane · 5 vs 0 public rows
Reading
Not comparable

Multilingual

Not comparable
GPT-6 Luna
Not ranked
MAI Code 2
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Instruction following

Not comparable
GPT-6 Luna
Not ranked
MAI Code 2
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Math

Not comparable
GPT-6 Luna
77.1
Unranked · 2 rankable rows
MAI Code 2
Not ranked
Basis
Provisional lane · 2 vs 0 weighted rows
Reading
Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, reasoning, multimodal, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.

Supported evidence per lane · bars run 0–100Methodology

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

GPT-6 Luna
$0.00035
Fits in one request
MAI Code 2
$0.0035
Fits in one request

GPT-6 Luna has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

GPT-6 Luna
$0.0065
Fits in one request
MAI Code 2
$0.065
Fits in one request

GPT-6 Luna has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

GPT-6 Luna
$0.009
Fits in one request
MAI Code 2
$0.08
Fits in one request

GPT-6 Luna has the lower modeled cost

Costs use the listed standard API rates.

Your workload

Enter a month of tokens to price it at both models’ listed API rates.

GPT-6 Luna
$2.00
per month
MAI Code 2
$20.00
per month

Standard listed rates, no batch or volume discounts.

Cached input falls back to the list input rate only where a cached rate is unpublished

Price against score, among all priced models

Overall score against output price per 1M tokens, one model per family. Pricing as of November 17, 2026.

GPT-6 Luna

20406080100$0.10$0.50$1$5$10$50$100Output price, $ per 1M tokens (log)Overall score
  • 1 model at the same price scores higher than GPT-6 Luna. The nearest is Claude Haiku 5.5 (score 62.9, also $0.50 per 1M output tokens).
  • MAI Code 2 has no overall score yet, so it is not plotted.

Gray dots are the other priced models. The dark step line is the efficiency frontier: for a model under it, another model costs no more and scores at least as high.

See GPT-6 Luna on the full price vs performance chart

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

Provider availability

GPT-6 Luna

Generally Available · OpenAI Responses API, OpenAI Chat Completions API, ChatGPT Work, Codex

OpenAI GPT-6 Sol and Luna launch

MAI Code 2

Not sourced

Reasoning profile

GPT-6 Luna

Reasoning

MAI Code 2

Reasoning

Weight access

GPT-6 Luna

Proprietary

MAI Code 2

Proprietary

License

GPT-6 Luna

Proprietary

MAI Code 2

Proprietary

Release date

GPT-6 Luna

2026-09-22

MAI Code 2

2026-11-17

If you already use one of these models

Deployment change
The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
Workload cost
Repository review: $0.0065 vs $0.065. Cache-heavy agent loop: $0.009 vs $0.08.
Context tradeoff
Both models list 1.05M.
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, GPT-6 Luna or MAI Code 2?

At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, GPT-6 Luna or MAI Code 2?

MAI Code 2 is not ranked on the public lane for coding, so no winner is named for coding.

Which is better for agentic tasks, GPT-6 Luna or MAI Code 2?

MAI Code 2 is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.

Which costs less, GPT-6 Luna or MAI Code 2?

For the stated presets, chat costs $0.00035 on GPT-6 Luna and $0.0035 on MAI Code 2; repository review costs $0.0065 and $0.065; the cache-heavy agent loop costs $0.009 and $0.08. Costs use the listed standard API rates.

Which has the larger context window, GPT-6 Luna or MAI Code 2?

Both models list the same context window, 1.05M.

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence31 rows

Agentic

  • Terminal-Bench 4.1

    GPT-6 Luna15.20%
    Source
    MAI Code 2—

    Not directly comparable

  • Agents' Last Exam (Linux)

    GPT-6 Luna57.1%
    Source
    MAI Code 2—

    Not directly comparable

  • ExploitGym

    GPT-6 Luna11.6%
    Source
    MAI Code 2—

    Not directly comparable

  • Hermes Index

    GPT-6 Luna33.9%
    Source
    MAI Code 2—

    Not directly comparable

  • Hermes Bench

    GPT-6 Luna70.5%
    Source
    MAI Code 2—

    Not directly comparable

  • MazeBench

    GPT-6 Luna0%
    Source
    MAI Code 2—

    Not directly comparable

  • OSWorld 2.1 Offline

    GPT-6 Luna48.9%
    Source
    MAI Code 2—

    Not directly comparable

  • Terminal-Bench 2.1 (Vals)

    GPT-6 Luna73.0%
    Source
    MAI Code 2—

    Not directly comparable

  • Terminal-Bench 2.1

    GPT-6 Luna—
    MAI Code 284.6%
    Source

    Not directly comparable

Coding

  • DeepSWE

    GPT-6 Luna66.6%
    Source
    MAI Code 273.2%
    Source

    MAI Code 2 leads this result

  • Bug Hunt Bench

    GPT-6 Luna18.3 fixes
    Source
    MAI Code 2—

    Not directly comparable

  • FrontierCode 2.0

    GPT-6 Luna22.7%
    Source
    MAI Code 2—

    Not directly comparable

  • VulcanBench Frontier v4

    GPT-6 Luna71.9%
    Source
    MAI Code 2—

    Not directly comparable

  • Terminal-Bench 2.1

    GPT-6 Luna—
    MAI Code 284.6%
    Source

    Not directly comparable

Reasoning

  • BioBasicsBench

    GPT-6 Luna37.6%
    Source
    MAI Code 2—

    Not directly comparable

  • ARC-AGI-1

    GPT-6 Luna86.67%
    Source
    MAI Code 2—

    Not directly comparable

  • ARC-AGI-2

    GPT-6 Luna59.3%
    Source
    MAI Code 2—

    Not directly comparable

  • ARC-AGI-3

    GPT-6 Luna0.1%
    Source
    MAI Code 2—

    Not directly comparable

  • ARC-AGI-3 Provider Adapter

    GPT-6 Luna0.59%
    Source
    MAI Code 2—

    Not directly comparable

Multimodal

  • ParseBench-2

    GPT-6 Luna74.46%
    Source
    MAI Code 2—

    Not directly comparable

  • PerceptionBench (tools)

    GPT-6 Luna62.8%
    Source
    MAI Code 2—

    Not directly comparable

  • ExtractBench Pro

    GPT-6 Luna0%
    Source
    MAI Code 2—

    Not directly comparable

  • ExtractBench (LlamaIndex)

    GPT-6 Luna91.01%
    Source
    MAI Code 2—

    Not directly comparable

  • ParseBench-1

    GPT-6 Luna64.13%
    Source
    MAI Code 2—

    Not directly comparable

Knowledge

  • HealthBench (raw)

    GPT-6 Luna50.0%
    Source
    MAI Code 2—

    Not directly comparable

  • HealthBench (length-adjusted)

    GPT-6 Luna54.5%
    Source
    MAI Code 2—

    Not directly comparable

  • HealthBench Professional

    GPT-6 Luna60.8%
    Source
    MAI Code 2—

    Not directly comparable

  • HealthBench Professional (raw)

    GPT-6 Luna61.2%
    Source
    MAI Code 2—

    Not directly comparable

  • HealthBench Hard

    GPT-6 Luna31.4%
    Source
    MAI Code 2—

    Not directly comparable

Math

  • FrontierMath v2 (Tiers 1-3)

    GPT-6 Luna78.947%
    Source
    MAI Code 2—

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    GPT-6 Luna56.098%
    Source
    MAI Code 2—

    Not directly comparable

31 public results · 1 shared

Watch GPT-6 Luna vs MAI Code 2

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 7,000+ subscribers.

Last updated November 17, 2026