Skip to main content
BenchLM
Data

Gemini 3 Pro vs Gemini 4 Argon

Updated November 13, 2026. Rank says Gemini 4 Argon is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.

Share or export
Share on XLinkedInAPI/MCP

Decision reading

Gemini 4 Argon has the higher public score, 79.03 versus 58.35, and the 90% score intervals do not overlap. 3 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 Pro

Google

58.35/100

Supported · Public rank #56

90% interval 47.0–69.7

Model B
Google logo
Gemini 4 Argon

Google

79.03/100

Supported · Public rank #8

90% interval 75.4–82.7

API price per 1M tokens

Gemini 3 Pro: $2 input / $12 output

Gemini 4 Argon: $2 input / $10 output

Context window

Gemini 3 Pro: 2M

Gemini 4 Argon: 1M

Released

Gemini 3 Pro: November 18, 2025

Gemini 4 Argon: September 30, 2026

Shared results
3
Gemini 3 Pro only
29
Gemini 4 Argon only
36
Like-for-like categories
2 / 8
Supported: Gemini 3 Pro and Gemini 4 ArgonHow 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.

  • Long documents

    Prompts that approach the documented context limit

    Gemini 3 Pro

    Gemini 3 Pro has the larger documented context window.

    Confidence: documented

  • Chat turn cost

    1K fresh input + 500 output tokens

    Gemini 4 Argon

    Gemini 4 Argon 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

    Gemini 4 Argon

    Gemini 4 Argon has the lower estimated token cost for this stated workload. Gemini 3 Pro has no published cached-input rate, so cached tokens use its listed input rate.

    Confidence: rate-fallback
Show secondary and unsupported calls

  • Repository review cost

    50K fresh input + 3K output tokens

    Gemini 4 Argon

    Gemini 4 Argon 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

    Gemini 3 Pro is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.

    Confidence: limited

  • Agentic work

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

    Not enough matched evidence

    Gemini 3 Pro is not ranked on the public lane for agentic, so no winner is named for agentic.

    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.

43.3Gemini 3 Pro71.8Gemini 4 Argon

Directional only · BenchAlign v5.8

Gemini 4 Argon has the higher coding point estimate. Conditional score ranges do not establish rank confidence.

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

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.

1 category rests 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.

Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.

Bars run 0–100 on each benchmark’s normalized display scale

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.

Reasoning

Like-for-like
Gemini 3 Pro
48.7
Supported · #54/160
Gemini 4 Argon
76.5
Supported · #7/160
Basis
BenchAlign v5.8 lane · 2 vs 7 public rows
Reading
Gemini 4 Argon leads

Knowledge

Like-for-like
Gemini 3 Pro
62.1
Supported · #40/180
Gemini 4 Argon
79.2
Supported · #8/180
Basis
BenchAlign v5.8 lane · 2 vs 2 public rows
Reading
Gemini 4 Argon leads · intervals overlap

Coding

Directional only
Gemini 3 Pro
43.3
Estimated · #69/162
Gemini 4 Argon
71.8
Supported · #9/162
Basis
BenchAlign v5.8 lane · 4 vs 9 public rows
Reading
Directional only

Agentic

Not comparable
Gemini 3 Pro
Not ranked
Gemini 4 Argon
74.9
Supported · #6/135
Basis
BenchAlign v5.8 lane · 2 vs 13 public rows
Reading
Not comparable

Multimodal

Not comparable
Gemini 3 Pro
58.7
Estimated · #40/67
Gemini 4 Argon
Not ranked
Basis
BenchAlign v5.8 lane · 22 vs 5 public rows
Reading
Not comparable

Multilingual

Not comparable
Gemini 3 Pro
Not ranked
Gemini 4 Argon
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Instruction following

Not comparable
Gemini 3 Pro
86.1
#48/157
Gemini 4 Argon
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Math

Not comparable
Gemini 3 Pro
Not ranked
Gemini 4 Argon
87.4
Unranked · 2 rankable rows
Basis
Provisional lane · 0 vs 2 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

Gemini 3 Pro
$0.008
Fits in one request
Gemini 4 Argon
$0.007
Fits in one request

Gemini 4 Argon has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

Gemini 3 Pro
$0.136
Fits in one request
Gemini 4 Argon
$0.13
Fits in one request

Gemini 4 Argon 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 Pro
$0.56
Fits in one request
Cached input priced at the published list-input rate
Gemini 4 Argon
$0.16
Fits in one request

Gemini 4 Argon has the lower modeled cost

Gemini 3 Pro has no published cached-input rate, so cached tokens use its listed input rate.

Your workload

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

Gemini 3 Pro
$44.00
per month
Gemini 4 Argon
$40.00
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 12, 2026.

Gemini 3 ProGemini 4 Argon

20406080100$0.10$0.50$1$5$10$50$100Output price, $ per 1M tokens (log)Overall score
  • 18 models score at least as high as Gemini 3 Pro for the same price or less. The nearest is Gemini 3.1 Pro (score 63.1, also $12.00 per 1M output tokens).
  • 2 models at the same price score higher than Gemini 4 Argon. The nearest is Gemini 4 Carbon (score 89.2, also $10.00 per 1M output tokens).

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 Gemini 4 Argon 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.

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

Gemini 3 Pro

Not published

Gemini 4 Argon

$0.1 per 1M cached input tokens

Google Gemini API pricing

Reasoning profile

Gemini 3 Pro

Non-Reasoning

Gemini 4 Argon

Reasoning

Weight access

Gemini 3 Pro

Proprietary

Gemini 4 Argon

Proprietary

License

Gemini 3 Pro

Proprietary

Gemini 4 Argon

Proprietary

Release date

Gemini 3 Pro

2025-11-18

Gemini 4 Argon

2026-09-30

If you already use one of these models

Deployment change
Both entries list Google as the provider. Confirm endpoint, model ID, limits, and feature support before switching.
Quality signal
Gemini 4 Argon has the higher public score, 79.03 versus 58.35, and the 90% score intervals do not overlap.
Workload cost
Repository review: $0.136 vs $0.13. Cache-heavy agent loop: $0.56 vs $0.16.
Context tradeoff
Gemini 3 Pro has the larger documented window (2M).
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, Gemini 3 Pro or Gemini 4 Argon?

Gemini 4 Argon has the higher public score, 79.03 versus 58.35, and the 90% score intervals do not overlap. That is the clearest overall quality signal in the current public data.

Which is better for coding, Gemini 3 Pro or Gemini 4 Argon?

Gemini 4 Argon scores higher for coding on the public lane, 71.8 to 43.3. Gemini 3 Pro is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, Gemini 3 Pro or Gemini 4 Argon?

Gemini 3 Pro is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.

Which costs less, Gemini 3 Pro or Gemini 4 Argon?

For the stated presets, chat costs $0.008 on Gemini 3 Pro and $0.007 on Gemini 4 Argon; repository review costs $0.136 and $0.13; the cache-heavy agent loop costs $0.56 and $0.16. Gemini 3 Pro has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, Gemini 3 Pro or Gemini 4 Argon?

Gemini 3 Pro has the larger documented context window: 2M, 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 evidence68 rows

Agentic

  • Gert Labs

    Gemini 3 Pro63.23%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • JobBench

    Gemini 3 Pro11.4%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • AutomationBench

    Gemini 3 Pro—
    Gemini 4 Argon51.3%
    Source

    Not directly comparable

  • Finance Agent v2

    Gemini 3 Pro—
    Gemini 4 Argon65.4%
    Source

    Not directly comparable

  • Terminal-Bench 4.0

    Gemini 3 Pro—
    Gemini 4 Argon57.40%
    Source

    Not directly comparable

  • Agents' Last Exam

    Gemini 3 Pro—
    Gemini 4 Argon39.5%
    Source

    Not directly comparable

  • OSWorld 2.0

    Gemini 3 Pro—
    Gemini 4 Argon69.2%
    Source

    Not directly comparable

  • CWE-bench v1

    Gemini 3 Pro—
    Gemini 4 Argon68.0%
    Source

    Not directly comparable

  • Terminal-Bench-Science 0.1 (6x verifier timeout)

    Gemini 3 Pro—
    Gemini 4 Argon57.6%
    Source

    Not directly comparable

  • MazeBench

    Gemini 3 Pro—
    Gemini 4 Argon12%
    Source

    Not directly comparable

  • OSWorld 2.1 Offline

    Gemini 3 Pro—
    Gemini 4 Argon74.7%
    Source

    Not directly comparable

  • ExploitGym

    Gemini 3 Pro—
    Gemini 4 Argon39.5%
    Source

    Not directly comparable

  • RoboDojo Sim

    Gemini 3 Pro—
    Gemini 4 Argon34.2%
    Source

    Not directly comparable

  • CWE-bench

    Gemini 3 Pro—
    Gemini 4 Argon65.0%
    Source

    Not directly comparable

  • Terminal-Bench 2.1 (Vals)

    Gemini 3 Pro—
    Gemini 4 Argon57.7%
    Source

    Not directly comparable

Coding

  • Vibe Code Bench

    Gemini 3 Pro14.30%
    Source
    Gemini 4 Argon91.90%
    Source

    Gemini 4 Argon leads this result

  • LiveCodeBench (Vals)

    Gemini 3 Pro86.4%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • SWE-bench (Vals)

    Gemini 3 Pro76.4%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • PostTrainBench v1

    Gemini 3 Pro18.1%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • ScarfBench

    Gemini 3 Pro—
    Gemini 4 Argon75.2%
    Source

    Not directly comparable

  • DeepSWE

    Gemini 3 Pro—
    Gemini 4 Argon77.9%
    Source

    Not directly comparable

  • FrontierSWE v2

    Gemini 3 Pro—
    Gemini 4 Argon55.0%
    Source

    Not directly comparable

  • PostTrainBench v1.1

    Gemini 3 Pro—
    Gemini 4 Argon45.3%
    Source

    Not directly comparable

  • Bug Hunt Bench

    Gemini 3 Pro—
    Gemini 4 Argon40.7 fixes
    Source

    Not directly comparable

  • FrontierCode 2.0

    Gemini 3 Pro—
    Gemini 4 Argon26.3%
    Source

    Not directly comparable

  • CursorBench 4.0

    Gemini 3 Pro—
    Gemini 4 Argon52.9%
    Source

    Not directly comparable

  • PostTrainBench

    Gemini 3 Pro—
    Gemini 4 Argon40.4%
    Source

    Not directly comparable

Reasoning

  • ARC-AGI-2

    Gemini 3 Pro31.1%
    Source
    Gemini 4 Argon91.3%
    Source

    Gemini 4 Argon leads this result

  • ARC-AGI-1

    Gemini 3 Pro75.00%
    Source
    Gemini 4 Argon98.50%
    Source

    Gemini 4 Argon leads this result

  • STEMBench

    Gemini 3 Pro—
    Gemini 4 Argon2.44
    Source

    Not directly comparable

  • Graphwalks BFS 128K

    Gemini 3 Pro—
    Gemini 4 Argon99.7%
    Source

    Not directly comparable

  • GraphWalks BFS 256K–1M

    Gemini 3 Pro—
    Gemini 4 Argon84.2%
    Source

    Not directly comparable

  • ARC-AGI-3

    Gemini 3 Pro—
    Gemini 4 Argon46.3%
    Source

    Not directly comparable

  • ARC-AGI-3 Provider Adapter

    Gemini 3 Pro—
    Gemini 4 Argon97.30%
    Source

    Not directly comparable

Multimodal

  • MMMU

    Gemini 3 Pro87.2%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • MMMU-Pro

    Gemini 3 Pro81%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • RealWorldQA

    Gemini 3 Pro83.3%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • OmniDocBench 1.5

    Gemini 3 Pro88.5%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • Video-MME (with subtitle)

    Gemini 3 Pro88.4%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • Video-MME (w/o subtitle)

    Gemini 3 Pro87.7%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • MathVision

    Gemini 3 Pro86.6%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • We-Math

    Gemini 3 Pro86.9%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • DynaMath

    Gemini 3 Pro85.1%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • SimpleVQA

    Gemini 3 Pro73.2%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • MMLongBench-Doc

    Gemini 3 Pro60.5%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • CC-OCR

    Gemini 3 Pro79.0%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • AI2D_TEST

    Gemini 3 Pro94.1%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • CountBench

    Gemini 3 Pro97.3%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • RefCOCO (avg)

    Gemini 3 Pro84.1%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • ODINW13

    Gemini 3 Pro46.3%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • ERQA

    Gemini 3 Pro70.5%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • VideoMMMU

    Gemini 3 Pro87.6%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • MLVU (M-Avg)

    Gemini 3 Pro83.0%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • ScreenSpot Pro

    Gemini 3 Pro72.7%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • CharXiv

    Gemini 3 Pro81.4%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • V*

    Gemini 3 Pro88.0%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • ParseBench-2

    Gemini 3 Pro—
    Gemini 4 Argon84.20%
    Source

    Not directly comparable

  • ExtractBench Pro

    Gemini 3 Pro—
    Gemini 4 Argon30%
    Source

    Not directly comparable

  • Chartography (no tools)

    Gemini 3 Pro—
    Gemini 4 Argon71.6%
    Source

    Not directly comparable

  • LVBench

    Gemini 3 Pro—
    Gemini 4 Argon91.7%
    Source

    Not directly comparable

  • Blueprint-Bench 2

    Gemini 3 Pro—
    Gemini 4 Argon54.4%
    Source

    Not directly comparable

Knowledge

  • GPQA Diamond (Vals)

    Gemini 3 Pro91.7%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • MMLU-Pro (Vals)

    Gemini 3 Pro90.1%
    Source
    Gemini 4 Argon—

    Not directly comparable

  • LABBench2

    Gemini 3 Pro—
    Gemini 4 Argon88.8%
    Source

    Not directly comparable

  • Morphology-to-molecule matching

    Gemini 3 Pro—
    Gemini 4 Argon25.2%
    Source

    Not directly comparable

Instruction following

  • Gray Swan IPI (15 attempts)

    Gemini 3 Pro—
    Gemini 4 Argon0.7%
    Source

    Not directly comparable

Math

  • FrontierMath v2 (Tiers 1-3)

    Gemini 3 Pro—
    Gemini 4 Argon85.965%
    Source

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    Gemini 3 Pro—
    Gemini 4 Argon78.049%
    Source

    Not directly comparable

68 public results · 3 shared

Related comparisons

Compare API pricingRead the methodologyOpen the model selector

Watch Gemini 3 Pro vs Gemini 4 Argon

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 13, 2026