UAB
Home/Compare
Compare
Live · updated continuously
Compare · pick a model

Compare the models people actually argue about.

Use shared tests, visible gaps, and close-call warnings before turning a benchmark lead into a public claim.
Selected models · 1
Shared benchmarks · 13
Current leader · Gemini 3.1 Pro
Compare controls

Keep the recommendation honest.

Use case and source filter stay visible up top. Edits, saves, follows, and links stay available without running the page.

Current frameBest for this use caseCoding copilot · All public sources
What should win?
Preset

Keep the same models, change the use case, and recompute the recommendation without resetting the page.

Sources
Models to compare

Keep the compare set tight. Add one model at a time, then prune aggressively.

1 / 6 models
Workspace utilities

Save, follow, or copy this exact setup only after the comparison looks right.

Gemini 3.1 Pro
Save or share this workspacefollow it, copy it, or reopen it later
Open reportOpen workspaces
Compare · Coding copilot · best for this use case

Gemini 3.1 Pro leads this compare set for coding copilot.

Gemini 3.1 Prono clear runner-up
Visible tradeoffs

The current evidence supports a shortlist, not a single winner.

Shared tests
13
Close calls
0
Sources
All public
Open evidence pageInspect disagreement
Decision detail

Read this page as a workspace for choosing between top models, not as a universal crown. The current frame is Coding copilot under best for this use case.

Why the leader is ahead
Reasoning / math / science · Long context
Where the evidence is thin
0 close-call benchmarks and 101 missing missing leader rows keep this from reading as settled.
What to do next
Inspect decisive benchmarks first, then open the disagreement page or head-to-head pages if the top line still feels too narrow.
Open evidence pageInspect benchmark disagreement
13 of 114 benchmarks
Search Arena
AR · rating
Search · Search / tool use
1,21086.7%
exact aliasverified runtime
Row details
Raw value
1,210
Percentile
86.7%
Last updated
recent
Eligibility
headline eligible
Identity
provider alias (0.92)
Source label
gemini-3.1-pro-grounding
Source row
n/a
TutorBench
SL · %
Text · Reasoning / math / science
53%20%
exact directverified runtime
Row details
Raw value
53%
Percentile
20%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
VTB
SL · %
Vision · Vision understanding
29%90.9%
exact directverified runtime
Row details
Raw value
29%
Percentile
90.9%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
PRBench Legal
SL · %
Text · Professional reasoning
44%41.7%
exact directverified runtime
Row details
Raw value
44%
Percentile
41.7%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
HiL-Bench
SL · %
Code · Coding
20.3%40%
exact directverified runtime
Row details
Raw value
20.3%
Percentile
40%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
MASK
SL · %
Text · Safety
42.4%0%
exact directverified runtime
Row details
Raw value
42.4%
Percentile
0%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
MultiNRC
SL · %
Text · Reasoning / math / science
64.7%100%
exact directverified runtime
Row details
Raw value
64.7%
Percentile
100%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
Humanity's Last Exam
OFF · %
Text · Reasoning / math / science
44.4%85.7%
Officialmanual verifiedmanual verified
Row details
Raw value
44.4%
Percentile
85.7%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
MRCR v2
OFF · %
Document · Long context
84.9%100%
Officialmanual verifiedmanual verified
Row details
Raw value
84.9%
Percentile
100%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
Terminal-Bench 2.0
OFF · %
Code · Coding
68.5%50%
Officialmanual verifiedmanual verified
Row details
Raw value
68.5%
Percentile
50%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
SWE-Bench Verified
OFF · %
Code · Coding
80.6%60%
Officialmanual verifiedmanual verified
Row details
Raw value
80.6%
Percentile
60%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
BrowseComp
OFF · %
Search · Search / tool use
85.9%100%
Officialmanual verifiedmanual verified
Row details
Raw value
85.9%
Percentile
100%
Last updated
recent
Eligibility
headline eligible
Identity
exact (1.00)
Source label
gemini-3-1-pro
Source row
n/a
1-12 of 13
Page 1 of 2
Decisive benchmarks

What is doing the visible work

No decisive benchmark
The shared tests are too split or too sparse for a clean separator.
Decision pressure

What changes the winner

Gemini 3.1 Pro
101 visible benchmark gaps still leave room for the result to move.
Coverage gaps

Where evidence is missing

Gemini 3.1 Pro
Missing visible evidence on this compare surface
101
Preset interpretation

How this weighting reads the field

The current decision mode is grounded in the Coding copilot preset. This keeps the compare page connected to a visible use case instead of an unspoken “overall winner” claim.

score = 1.45 × coding + 1.05 × reasoning math science + 0.80 × long contextcoverage floor = 55% · recency window = 120 days
Gemini 3.1 Pro
Google
62.8%
Share or publish this compare resultlinks, copy, and advanced framings
Share this report

Share the recommendation with the source data attached.

Use the report for the full source trail, or the card image when the post needs a clean preview.

Model compareGemini 3.1 Pro leads this compare set for coding copilot.

0 shared benchmarks are still too close to call, so the win stays conditional. This compare uses all public sources, with provider-official evidence labeled separately.

Report/artifacts/compare?models=gemini-3-1-pro&mode=best-for-this-use-case&preset=coding-copilot
Card image/api/og?kind=compare&models=gemini-3-1-pro&mode=best-for-this-use-case&preset=coding-copilot
Public links

Open or copy the stable surfaces

The report is the canonical source page. The card image is the compact preview for embeds, screenshots, and social cards.

Open reportOpen card preview
Copy-ready text

Use the exact public framing

Each copy action keeps the recommendation attached to source data instead of forcing you into a blank composer.

Advanced framings and X composerNeutral, contrarian, open-model, and skeptical variants
Model compare

Pick the voice before you post

Use the framing variants only when you need them. The report and the public copy actions above should handle most cases.

Neutral analystLead with the recommendation, then attach the reason and warning.Gemini 3.1 Pro leads this compare set for coding copilot.
ContrarianPush against the easy read and keep the strongest alternative live.Contrarian take: Gemini 3.1 Pro leads this compare set for coding copilot.
Open-model angleBias the framing toward open weights and source transparency.Open-model angle: Model compare · Gemini 3.1 Pro
Don't trust the headlineLead with the warning before you let the claim travel.Don't trust the headline: Model compare · Gemini 3.1 Pro
X composer

Compose a post that keeps the warning attached

The post shell always exposes the recommendation, why, warning, source link, and an optional discussion question.

HeadlineGemini 3.1 Pro leads this compare set for coding copilot.
WhyThe visible source data changed enough to change the recommendation.
Warning0 shared benchmarks are still too close to call, so the win stays conditional. This compare uses all public sources, with provider-official evidence labeled separately.
Source link/artifacts/compare?models=gemini-3-1-pro&mode=best-for-this-use-case&preset=coding-copilot
PreviewOver 280
Gemini 3.1 Pro leads this compare set for coding copilot.
The visible source data changed enough to change the recommendation.
Warning: 0 shared benchmarks are still too close to call, so the win stays conditional. This compare uses all public sources, with provider-official evidence labeled separately.
Source data: /artifacts/compare?models=gemini-3-1-pro&mode=best-for-this-use-case&preset=coding-copilot
Open in XOpen card preview
Reading guide

How to read this workspace

Who wins most oftenBenchmarks with one clear percentile leader.Missing evidenceBenchmarks where a top model has no visible score.Decision readA public claim tied to an explicit use case and source filter.