Skip to main content
BenchLM
Advertise

Model comparison

Gemini 3 Pro Deep Think vs Grok 4.20

Updated July 29, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Gemini 3 Pro Deep Think

Google

60.5/100

Estimated · Public rank #45

90% interval 49.0–72.0

Grok 4.20

xAI

53.9/100

Estimated · Public rank #93

90% interval 37.1–70.6

Gemini 3 Pro Deep Think has the higher public score estimate, 60.5 versus 53.88, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

1 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.

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.

No workload recommendation clears the current evidence threshold.

Use the matched evidence, workload costs, and sourced specifications below instead of treating a point score as a universal answer.

Show secondary and unsupported calls

  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Agentic work

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

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Long documents

    Prompts that approach the documented context limit

    Not enough matched evidence

    A complete context comparison is not sourced.

    Confidence: limited

  • Chat turn cost

    1K fresh input + 500 output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

  • Cache-heavy agent loop cost

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

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: rate-fallback

  • Repository review cost

    50K fresh input + 3K output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

What is actually comparable

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

Shared results
1
Gemini 3 Pro Deep Think only
0
Grok 4.20 only
17
Like-for-like categories
1 / 8

Category results, on a stated basis

Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.

Reasoning

Like-for-like
Gemini 3 Pro Deep Think
45.1
Grok 4.20
53.3
Weighted basis
1 vs 1 rows
Reading
Grok 4.20 leads

Agentic

Not comparable
Gemini 3 Pro Deep Think
Not measured
Grok 4.20
47.1
Weighted basis
0 vs 1 rows
Reading
Not comparable

Coding

Not comparable
Gemini 3 Pro Deep Think
Not measured
Grok 4.20
67.1
Weighted basis
0 vs 2 rows
Reading
Not comparable

Knowledge

Not comparable
Gemini 3 Pro Deep Think
Not measured
Grok 4.20
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Math

Not comparable
Gemini 3 Pro Deep Think
Not measured
Grok 4.20
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
Gemini 3 Pro Deep Think
Not measured
Grok 4.20
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multimodal

Not comparable
Gemini 3 Pro Deep Think
Not measured
Grok 4.20
70.1
Weighted basis
0 vs 2 rows
Reading
Not comparable

Instruction following

Not comparable
Gemini 3 Pro Deep Think
Not measured
Grok 4.20
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

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.

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 Deep Think
API rate not published
Fit state unavailable
Grok 4.20
$0.005
Fits in one request

Gemini 3 Pro Deep Think has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

Gemini 3 Pro Deep Think
API rate not published
Fit state unavailable
Grok 4.20
$0.118
Fits in one request

Gemini 3 Pro Deep Think has no comparable published API token rate.

Cache-heavy agent loop

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

Gemini 3 Pro Deep Think
API rate not published
Fit state unavailable
Cached-input rate unavailable
Grok 4.20
$0.5
Fits in one request
Cached input priced at the published list-input rate

Grok 4.20 has no published cached-input rate, so cached tokens use its listed input rate. Gemini 3 Pro Deep Think has no comparable published API token rate.

Specification differences

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

API model ID

Gemini 3 Pro Deep Think

Not sourced

Grok 4.20

Not sourced

Cached-input rate

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

Gemini 3 Pro Deep Think

No comparable hosted API rate

Grok 4.20

Not published

Documented inputs

Gemini 3 Pro Deep Think

Not sourced

Grok 4.20

Not sourced

Documented outputs

Gemini 3 Pro Deep Think

Not sourced

Grok 4.20

Not sourced

Provider availability

Gemini 3 Pro Deep Think

Limited Access · Gemini app for Google AI Ultra, Gemini API early-access program

Google Gemini 3 Deep Think launch

Grok 4.20

Not sourced

Reasoning profile

Gemini 3 Pro Deep Think

Reasoning

Grok 4.20

Reasoning

Weight access

Gemini 3 Pro Deep Think

Proprietary

Grok 4.20

Proprietary

License

Gemini 3 Pro Deep Think

Proprietary

Grok 4.20

Proprietary

Release date

Gemini 3 Pro Deep Think

2026-02-12

Grok 4.20

2026-03-10

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
Gemini 3 Pro Deep Think has the higher public score estimate, 60.5 versus 53.88, but the 90% score intervals overlap.
Workload cost
A complete comparable API-rate estimate is not available for both models.
Context tradeoff
A complete documented context comparison is not available.

Run the same representative tasks against both endpoints before changing production traffic.

Benchmark evidence

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

Browse raw public benchmark evidence18 rows

Agentic

  • Terminal-Bench 2.0

    Gemini 3 Pro Deep Think—
    Grok 4.2047.1%
    Source

    Not directly comparable

  • DeepSearchQA

    Gemini 3 Pro Deep Think—
    Grok 4.2062.8%
    Source

    Not directly comparable

  • Gert Labs

    Gemini 3 Pro Deep Think—
    Grok 4.2038.36%
    Source

    Not directly comparable

Coding

  • LiveCodeBench Pro

    Gemini 3 Pro Deep Think—
    Grok 4.2074.2%
    Source

    Not directly comparable

  • SWE-bench Verified

    Gemini 3 Pro Deep Think—
    Grok 4.2076.7%
    Source

    Not directly comparable

  • SWE-bench Pro

    Gemini 3 Pro Deep Think—
    Grok 4.2051.8%
    Source

    Not directly comparable

  • Vibe Code Bench

    Gemini 3 Pro Deep Think—
    Grok 4.204.06%
    Source

    Not directly comparable

Reasoning

  • ARC-AGI-2

    Gemini 3 Pro Deep Think45.1%
    Source
    Grok 4.2053.3%
    Source

    Grok 4.20 leads this result

  • ARC-AGI-3

    Gemini 3 Pro Deep Think—
    Grok 4.200.1%
    Source

    Not directly comparable

Knowledge

  • GPQA-D

    Gemini 3 Pro Deep Think—
    Grok 4.2088.5%
    Source

    Not directly comparable

  • HLE w/o tools

    Gemini 3 Pro Deep Think—
    Grok 4.2031.6%
    Source

    Not directly comparable

  • HealthBench Hard

    Gemini 3 Pro Deep Think—
    Grok 4.2020.3%
    Source

    Not directly comparable

  • MedXpertQA (Text)

    Gemini 3 Pro Deep Think—
    Grok 4.2050.2%
    Source

    Not directly comparable

Multimodal

  • MMMU-Pro

    Gemini 3 Pro Deep Think—
    Grok 4.2075.2%
    Source

    Not directly comparable

  • CharXiv

    Gemini 3 Pro Deep Think—
    Grok 4.2060.9%
    Source

    Not directly comparable

  • ERQA

    Gemini 3 Pro Deep Think—
    Grok 4.2054.1%
    Source

    Not directly comparable

  • SimpleVQA

    Gemini 3 Pro Deep Think—
    Grok 4.2057.4%
    Source

    Not directly comparable

  • MedXpertQA (MM)

    Gemini 3 Pro Deep Think—
    Grok 4.2065.8%
    Source

    Not directly comparable

Frequently asked questions

Which is better, Gemini 3 Pro Deep Think or Grok 4.20?

Gemini 3 Pro Deep Think has the higher public score estimate, 60.5 versus 53.88, 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 Pro Deep Think or Grok 4.20?

The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.

Which is better for agentic tasks, Gemini 3 Pro Deep Think or Grok 4.20?

The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.

Which costs less, Gemini 3 Pro Deep Think or Grok 4.20?

Both models do not have comparable published API token rates, so this page does not name a universal price winner.

Which has the larger context window, Gemini 3 Pro Deep Think or Grok 4.20?

A complete documented context-window comparison is not available.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated July 29, 2026

Keep the decision current

One weekly note on benchmark changes, pricing moves, and models worth re-testing.