Skip to main content
BenchLM

Gemini 3.1 Pro vs GPT-6 Luna

Updated September 23, 2026. Rank says GPT-6 Luna is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.

Share or export
Share on XLinkedInSocial cardCSVJSON

Decision reading

GPT-6 Luna has the higher public score estimate, 66.52 versus 64.97, but the 90% score intervals overlap. Treat that as a lead, not a settled 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
Google logo
Gemini 3.1 Pro

Google

64.97/100

Estimated · Public rank #29

90% interval 58.9–71.1

Model B
OpenAI logo
GPT-6 Luna

OpenAI

66.52/100

Estimated · Public rank #21

90% interval 55.0–78.0

Shared results
1
Gemini 3.1 Pro only
27
GPT-6 Luna only
6
Like-for-like categories
1 / 8
Estimated: Gemini 3.1 Pro and 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.

  • Coding work

    Code generation, repair, and software-engineering tasks

    GPT-6 Luna

    GPT-6 Luna leads on the public coding lane, 53.9 to 43.4, with Supported evidence for both models, although the 90% intervals overlap.

    Confidence: limited

  • Long documents

    Prompts that approach the documented context limit

    GPT-6 Luna

    GPT-6 Luna has the larger documented context window.

    Confidence: documented

  • 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
Show secondary and unsupported calls

  • 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

  • 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

  • Agentic work

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

    Not enough matched evidence

    GPT-6 Luna is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.

    Confidence: limited

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.

43.4Gemini 3.1 Pro53.9GPT-6 Luna

Like-for-like · BenchAlign v5.6

GPT-6 Luna leads the like-for-like coding row, although the 90% intervals overlap.

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

Open the coding leaderboardShortlist models for this job
Same basis rules as the category table below

What is actually comparable

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

2 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.

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 v5.6 lane for agentic, coding, 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.

Coding

Like-for-like
Gemini 3.1 Pro
43.4
Supported · #53/135
GPT-6 Luna
53.9
Supported · #32/135
Basis
BenchAlign v5.6 lane · 5 vs 1 public rows
Reading
GPT-6 Luna leads · intervals overlap

Agentic

Directional only
Gemini 3.1 Pro
38.9
Supported · #47/105
GPT-6 Luna
55.4
Estimated · #26/105
Basis
BenchAlign v5.6 lane · 6 vs 1 public rows
Reading
Directional only

Knowledge

Directional only
Gemini 3.1 Pro
64.2
Supported · #23/160
GPT-6 Luna
66.3
Estimated · #18/160
Basis
BenchAlign v5.6 lane · 6 vs 5 public rows
Reading
Directional only

Reasoning

Not comparable
Gemini 3.1 Pro
50.7
Unranked · 2 rankable rows
GPT-6 Luna
78.3
#6/18
Basis
Provisional lane · 2 vs 0 weighted rows
Reading
Not comparable

Multimodal

Not comparable
Gemini 3.1 Pro
79.1
#13/50
GPT-6 Luna
71.3
Unranked · 1 rankable row
Basis
Provisional lane · 2 vs 0 weighted rows
Reading
Not comparable

Multilingual

Not comparable
Gemini 3.1 Pro
Not ranked
GPT-6 Luna
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Instruction following

Not comparable
Gemini 3.1 Pro
Not ranked
GPT-6 Luna
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Math

Not comparable
Gemini 3.1 Pro
54.2
Unranked · 2 rankable rows
GPT-6 Luna
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, and knowledge denominators (BenchAlign v5.6) 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

Gemini 3.1 Pro
$0.008
Fits in one request
GPT-6 Luna
$0.00035
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

Gemini 3.1 Pro
$0.136
Fits in one request
GPT-6 Luna
$0.0065
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

Gemini 3.1 Pro
$0.2
Fits in one request
GPT-6 Luna
$0.009
Fits in one request

GPT-6 Luna has the lower modeled cost

Costs use the listed standard API rates.

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

Specification differences

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

Reasoning profile

Gemini 3.1 Pro

Reasoning

GPT-6 Luna

Reasoning

Weight access

Gemini 3.1 Pro

Proprietary

GPT-6 Luna

Proprietary

License

Gemini 3.1 Pro

Proprietary

GPT-6 Luna

Proprietary

Release date

Gemini 3.1 Pro

2026-02-19

GPT-6 Luna

2026-09-16

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
GPT-6 Luna has the higher public score estimate, 66.52 versus 64.97, but the 90% score intervals overlap.
Workload cost
Repository review: $0.136 vs $0.0065. Cache-heavy agent loop: $0.2 vs $0.009.
Context tradeoff
GPT-6 Luna has the larger documented window (1.05M).
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, Gemini 3.1 Pro or GPT-6 Luna?

GPT-6 Luna has the higher public score estimate, 66.52 versus 64.97, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, Gemini 3.1 Pro or GPT-6 Luna?

GPT-6 Luna leads the public coding lane, 53.9 to 43.4, with Supported evidence for both models, although the 90% intervals overlap.

Which is better for agentic tasks, Gemini 3.1 Pro or GPT-6 Luna?

GPT-6 Luna scores higher for agentic tasks on the public lane, 55.4 to 38.9. GPT-6 Luna is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which costs less, Gemini 3.1 Pro or GPT-6 Luna?

For the stated presets, chat costs $0.008 on Gemini 3.1 Pro and $0.00035 on GPT-6 Luna; repository review costs $0.136 and $0.0065; the cache-heavy agent loop costs $0.2 and $0.009. Costs use the listed standard API rates.

Which has the larger context window, Gemini 3.1 Pro or GPT-6 Luna?

GPT-6 Luna has the larger documented context window: 1.05M, compared with 1M.

Benchmark evidence

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

Browse raw public benchmark evidence34 rows

Agentic

  • Claw-Eval

    Gemini 3.1 Pro57.8%
    Source
    GPT-6 Luna—

    Not directly comparable

  • DeepSearchQA

    Gemini 3.1 Pro69.7%
    Source
    GPT-6 Luna—

    Not directly comparable

  • τ²-bench results

    Gemini 3.1 Pro95.6%
    Source
    GPT-6 Luna—

    Not directly comparable

  • Gert Labs

    Gemini 3.1 Pro56.87%
    Source
    GPT-6 Luna—

    Not directly comparable

  • ResearchClawBench

    Gemini 3.1 Pro13.3%
    Source
    GPT-6 Luna—

    Not directly comparable

  • Terminal-Bench 2.1 (Vals)

    Gemini 3.1 Pro70.8%
    Source
    GPT-6 Luna—

    Not directly comparable

  • ExploitGym

    Gemini 3.1 Pro—
    GPT-6 Luna11.6%
    Source

    Not directly comparable

Coding

  • LiveCodeBench Pro

    Gemini 3.1 Pro82.9%
    Source
    GPT-6 Luna—

    Not directly comparable

  • React Native Evals

    Gemini 3.1 Pro78.9%
    Source
    GPT-6 Luna—

    Not directly comparable

  • Vibe Code Bench

    Gemini 3.1 Pro32.03%
    Source
    GPT-6 Luna—

    Not directly comparable

  • LiveCodeBench (Vals)

    Gemini 3.1 Pro88.5%
    Source
    GPT-6 Luna—

    Not directly comparable

  • SWE-bench (Vals)

    Gemini 3.1 Pro78.8%
    Source
    GPT-6 Luna—

    Not directly comparable

  • DeepSWE

    Gemini 3.1 Pro—
    GPT-6 Luna66.6%
    Source

    Not directly comparable

Reasoning

  • ARC-AGI-2

    Gemini 3.1 Pro77.1%
    Source
    GPT-6 Luna—

    Not directly comparable

  • ARC-AGI-3

    Gemini 3.1 Pro0.4%
    Source
    GPT-6 Luna—

    Not directly comparable

Multimodal

  • MMMU-Pro

    Gemini 3.1 Pro83.9%
    Source
    GPT-6 Luna—

    Not directly comparable

  • CharXiv

    Gemini 3.1 Pro80.2%
    Source
    GPT-6 Luna—

    Not directly comparable

  • ERQA

    Gemini 3.1 Pro69.4%
    Source
    GPT-6 Luna—

    Not directly comparable

  • SimpleVQA

    Gemini 3.1 Pro72.4%
    Source
    GPT-6 Luna—

    Not directly comparable

  • ScreenSpot Pro

    Gemini 3.1 Pro84.4%
    Source
    GPT-6 Luna—

    Not directly comparable

  • ZeroBench

    Gemini 3.1 Pro29.0%
    Source
    GPT-6 Luna—

    Not directly comparable

  • MedXpertQA (MM)

    Gemini 3.1 Pro81.3%
    Source
    GPT-6 Luna—

    Not directly comparable

Knowledge

  • GPQA-D

    Gemini 3.1 Pro94.3%
    Source
    GPT-6 Luna—

    Not directly comparable

  • HLE w/o tools

    Gemini 3.1 Pro45.4%
    Source
    GPT-6 Luna—

    Not directly comparable

  • HealthBench Hard

    Gemini 3.1 Pro20.6%
    Source
    GPT-6 Luna31.4%
    Source

    GPT-6 Luna leads this result

  • MedXpertQA (Text)

    Gemini 3.1 Pro71.5%
    Source
    GPT-6 Luna—

    Not directly comparable

  • GPQA Diamond (Vals)

    Gemini 3.1 Pro95.5%
    Source
    GPT-6 Luna—

    Not directly comparable

  • MMLU-Pro (Vals)

    Gemini 3.1 Pro91.0%
    Source
    GPT-6 Luna—

    Not directly comparable

  • HealthBench (raw)

    Gemini 3.1 Pro—
    GPT-6 Luna50.0%
    Source

    Not directly comparable

  • HealthBench (length-adjusted)

    Gemini 3.1 Pro—
    GPT-6 Luna54.5%
    Source

    Not directly comparable

  • HealthBench Professional

    Gemini 3.1 Pro—
    GPT-6 Luna60.8%
    Source

    Not directly comparable

  • HealthBench Professional (raw)

    Gemini 3.1 Pro—
    GPT-6 Luna61.2%
    Source

    Not directly comparable

Math

  • FrontierMath v2 (Tiers 1-3)

    Gemini 3.1 Pro36.900%
    Source
    GPT-6 Luna—

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    Gemini 3.1 Pro16.700%
    Source
    GPT-6 Luna—

    Not directly comparable

34 public results · 1 shared

Related comparisons

Compare API pricingRead the methodologyOpen the model selector

Watch Gemini 3.1 Pro vs GPT-6 Luna

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

Read a sample issue

Join 2,000+ readers.

Last updated September 23, 2026