Enter opens the highlighted result. Esc closes.

Gemini 2.5 Pro vs Gemini 4 Neon

Gemini 4 Neon leads Gemini 2.5 Pro on coding, 86.9 to 19.2 on the Vibe Coding Index, and Gemini 2.5 Pro costs 1.5x less per output token ($10.00 vs $15.00 per million). Updated November 22, 2026; 38 benchmarks have a result for both.

Gemini 2.5 ProGoogle
Coding
19.2
Overall
46.6
Output $/M
$10.00
0
Gemini 4 NeonGoogle
Coding
86.9
Overall
89.1
Output $/M
$15.00
38
vs boards won of 38 shared

Gemini 2.5 Pro or Gemini 4 Neon: which to pick

Pick Gemini 4 Neon for coding: it scores 86.9 on the Vibe Coding Index against 19.2 for Gemini 2.5 Pro.

Pick Gemini 2.5 Pro for output tokens 1.5x cheaper ($10.00 vs $15.00 per million).

Shared results
38
Gemini 2.5 Pro only
54
Gemini 4 Neon only
31
Like-for-like categories
7 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Gemini 2.5 Pro 19.2 from 3 benchmarks, Gemini 4 Neon 86.9 from 4 benchmarks.
Widest category
Agentic work, 96.4 points apart on average over 1 shared benchmark.
Widest single gap
FrontierMath, tier 4: Gemini 2.5 Pro 0.0% against Gemini 4 Neon 97.6%, 97.6 points apart on a 0 to 100 scale.
Output price
Gemini 2.5 Pro $10.00 against Gemini 4 Neon $15.00 per million output tokens, 1.5x apart.
Context window
Both take 1.05M tokens.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryGemini 2.5 ProBarsGemini 4 NeonTies
Coding06
Agentic work01
Reasoning04
Math05
Knowledge04
Web design05
General013
All shared boards038
CodingLike-for-likeSix of these boards feed the Vibe Coding Index06
Coding benchmarks for Gemini 2.5 Pro and Gemini 4 Neon
BenchmarkGemini 2.5 ProGemini 4 NeonLeader
SWE-bench Verified53.6%Not measuredOnly Gemini 2.5 Pro measured
LiveBench codingNot measured89.7%Only Gemini 4 Neon measured
LMArena WebDev12271765Gemini 4 Neon
SWE-bench Verified (Vals AI)54.4%Not measuredOnly Gemini 2.5 Pro measured
Terminal-Bench 4 (Vals AI)Not measured61.6%Only Gemini 4 Neon measured
Vibe Code Bench (Vals AI)0.4%93.7%Gemini 4 Neon
LiveCodeBench (Artificial Analysis run)80.1%Not measuredOnly Gemini 2.5 Pro measured
SciCode46.3%Not measuredOnly Gemini 2.5 Pro measured
Terminal-Bench Hard (Artificial Analysis run)26.5%Not measuredOnly Gemini 2.5 Pro measured
Terminal-Bench 4.0 (Artificial Analysis run)0.0%63.6%Gemini 4 Neon
FrontierCode 2.0Not measured44.4%Only Gemini 4 Neon measured
Terminal-Bench32.6%Not measuredOnly Gemini 2.5 Pro measured
GSO-Bench3.9%Not measuredOnly Gemini 2.5 Pro measured
WeirdML54.0%Not measuredOnly Gemini 2.5 Pro measured
LMArena Text: coding14521555Gemini 4 Neon
LMArena Text: software and IT14591542Gemini 4 Neon
LMArena WebDev: ReactNot measured1787Only Gemini 4 Neon measured
LMArena WebDev: HTML12351803Gemini 4 Neon
LMArena WebDev: from an image1285Not measuredOnly Gemini 2.5 Pro measured
GameBench 2Not measured98.0Only Gemini 4 Neon measured
Terminal-Bench 2.032.6%Not measuredOnly Gemini 2.5 Pro measured
CursorBench 4.0Not measured65.4%Only Gemini 4 Neon measured
Neither model has a result on the other 30 boards here
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
Intelligence benchmarks for Gemini 2.5 Pro and Gemini 4 Neon
BenchmarkGemini 2.5 ProGemini 4 NeonLeader
AA Intelligence IndexNot measured61.4Only Gemini 4 Neon measured
Epoch Capabilities Index144.9Not measuredOnly Gemini 2.5 Pro measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for Gemini 2.5 Pro and Gemini 4 Neon
BenchmarkGemini 2.5 ProGemini 4 NeonLeader
LiveBench researchNot measured94.5%Only Gemini 4 Neon measured
LiveBench browserNot measured95.6%Only Gemini 4 Neon measured
LiveBench long agenticNot measured82.6%Only Gemini 4 Neon measured
Tax Agent Bench (Vals AI)Not measured54.9%Only Gemini 4 Neon measured
Legal Research Bench (Vals AI)Not measured56.3%Only Gemini 4 Neon measured
Harvey's Legal Agent Benchmark (Vals AI)Not measured19.6%Only Gemini 4 Neon measured
BioMysteryBench (Vals AI)Not measured80.0%Only Gemini 4 Neon measured
Terminal-Bench-Science (Vals AI)Not measured62.9%Only Gemini 4 Neon measured
Vals Science IndexNot measured62.6%Only Gemini 4 Neon measured
τ²-Bench Telecom54.1%Not measuredOnly Gemini 2.5 Pro measured
APEX-AgentsNot measured82.9%Only Gemini 4 Neon measured
GDPval23.3%Not measuredOnly Gemini 2.5 Pro measured
DeepResearch Bench41.5%Not measuredOnly Gemini 2.5 Pro measured
Vending-Bench 2$574$16,843Gemini 4 Neon
LMArena Search1143Not measuredOnly Gemini 2.5 Pro measured
BrowseComp-Plus29.5%Not measuredOnly Gemini 2.5 Pro measured
τ³-Banking13.7%Not measuredOnly Gemini 2.5 Pro measured
Neither model has a result on the other 27 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index04
Reasoning benchmarks for Gemini 2.5 Pro and Gemini 4 Neon
BenchmarkGemini 2.5 ProGemini 4 NeonLeader
LiveBench reasoningNot measured89.6%Only Gemini 4 Neon measured
MysteryMechanism (Vals AI)Not measured45.0%Only Gemini 4 Neon measured
GPQA Diamond (Artificial Analysis run)84.4%Not measuredOnly Gemini 2.5 Pro measured
Humanity's Last Exam (Artificial Analysis run)22.5%Not measuredOnly Gemini 2.5 Pro measured
HLE-Diamond (Artificial Analysis run)Not measured71.3%Only Gemini 4 Neon measured
CritPt v2, hard subset (Artificial Analysis run)Not measured61.5%Only Gemini 4 Neon measured
GPQA Diamond (Epoch AI run)85.3%96.3%Gemini 4 Neon
ARC-AGI41.0%Not measuredOnly Gemini 2.5 Pro measured
ARC-AGI-24.9%Not measuredOnly Gemini 2.5 Pro measured
SimpleBenchNot measured94.4%Only Gemini 4 Neon measured
Chess puzzlesNot measured83.0%Only Gemini 4 Neon measured
Mystery game puzzlesNot measured85.0%Only Gemini 4 Neon measured
EBR-benchNot measured98.1%Only Gemini 4 Neon measured
LMArena Text: hard prompts14551547Gemini 4 Neon
LMArena Text: expert14531560Gemini 4 Neon
LMArena Text: hard prompts in English14521524Gemini 4 Neon
LMArena Vision: diagrams1273Not measuredOnly Gemini 2.5 Pro measured
LMArena Vision: homework1282Not measuredOnly Gemini 2.5 Pro measured
Kagi LLM Benchmark70.3%Not measuredOnly Gemini 2.5 Pro measured
Neither model has a result on the other 7 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index05
Math benchmarks for Gemini 2.5 Pro and Gemini 4 Neon
BenchmarkGemini 2.5 ProGemini 4 NeonLeader
AIME 2025 (Artificial Analysis run)87.7%Not measuredOnly Gemini 2.5 Pro measured
OTIS Mock AIME84.7%100.0%Gemini 4 Neon
FrontierMath, tiers 1 to 324.6%91.2%Gemini 4 Neon
FrontierMath, tier 40.0%97.6%Gemini 4 Neon
LMArena Text: math14501548Gemini 4 Neon
LMArena Text: mathematical work14551566Gemini 4 Neon
MathArena11.7%Not measuredOnly Gemini 2.5 Pro measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index04
Knowledge benchmarks for Gemini 2.5 Pro and Gemini 4 Neon
BenchmarkGemini 2.5 ProGemini 4 NeonLeader
CaseLaw (Vals AI)63.9%Not measuredOnly Gemini 2.5 Pro measured
CorpFin (Vals AI)60.8%Not measuredOnly Gemini 2.5 Pro measured
MortgageTax (Vals AI)68.9%Not measuredOnly Gemini 2.5 Pro measured
MedCode (Vals AI)50.6%Not measuredOnly Gemini 2.5 Pro measured
MMLU Pro (Artificial Analysis run)86.2%Not measuredOnly Gemini 2.5 Pro measured
AA-OmniscienceNot measured54.8Only Gemini 4 Neon measured
SimpleQA VerifiedNot measured75.2%Only Gemini 4 Neon measured
LMArena Text: science14801549Gemini 4 Neon
LMArena Text: medicine and healthcare14761534Gemini 4 Neon
LMArena Text: legal and government14801518Gemini 4 Neon
LMArena Text: business and finance14401493Gemini 4 Neon
PRBench Finance38.9%Not measuredOnly Gemini 2.5 Pro measured
PRBench Legal38.9%Not measuredOnly Gemini 2.5 Pro measured
Neither model has a result on the other 8 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index05
Web design benchmarks for Gemini 2.5 Pro and Gemini 4 Neon
BenchmarkGemini 2.5 ProGemini 4 NeonLeader
Design Arena: websites11541347Gemini 4 Neon
Design Arena: UI components11231349Gemini 4 Neon
Design Arena: game dev10831394Gemini 4 Neon
Design Arena: data viz1216Not measuredOnly Gemini 2.5 Pro measured
Design Arena: 3D10471490Gemini 4 Neon
Design Arena: all code categories11341391Gemini 4 Neon
LiveBench designNot measured81.5%Only Gemini 4 Neon measured
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index013
General benchmarks for Gemini 2.5 Pro and Gemini 4 Neon
BenchmarkGemini 2.5 ProGemini 4 NeonLeader
Vals IndexNot measured70.9%Only Gemini 4 Neon measured
LiveBench data analysisNot measured85.3%Only Gemini 4 Neon measured
MedScribe (Vals AI)73.5%Not measuredOnly Gemini 2.5 Pro measured
IFBench48.7%Not measuredOnly Gemini 2.5 Pro measured
Long context reasoning (AA-LCR)69.0%Not measuredOnly Gemini 2.5 Pro measured
GDPval-AA463Not measuredOnly Gemini 2.5 Pro measured
DTBench82.4%Not measuredOnly Gemini 2.5 Pro measured
LMCA34.8%Not measuredOnly Gemini 2.5 Pro measured
Furniture assemblyNot measured89.2%Only Gemini 4 Neon measured
LMArena Text14581524Gemini 4 Neon
LMArena Text: instruction following14371535Gemini 4 Neon
LMArena Text: creative writing14541519Gemini 4 Neon
LMArena Vision1263Not measuredOnly Gemini 2.5 Pro measured
LMArena Text: English14581511Gemini 4 Neon
LMArena Text: non-English14521524Gemini 4 Neon
LMArena Text: Chinese15071598Gemini 4 Neon
LMArena Text: German1487Not measuredOnly Gemini 2.5 Pro measured
LMArena Text: Japanese1461Not measuredOnly Gemini 2.5 Pro measured
LMArena Text: Korean1434Not measuredOnly Gemini 2.5 Pro measured
LMArena Text: French14741563Gemini 4 Neon
LMArena Text: Spanish14741505Gemini 4 Neon
LMArena Text: Russian14611554Gemini 4 Neon
LMArena Text: Polish1479Not measuredOnly Gemini 2.5 Pro measured
LMArena Text: multi-turn14521525Gemini 4 Neon
LMArena Text: longer queries14491533Gemini 4 Neon
LMArena Text: writing and language14541517Gemini 4 Neon
LMArena Text: entertainment and media14411495Gemini 4 Neon
LMArena Vision: English1260Not measuredOnly Gemini 2.5 Pro measured
LMArena Vision: Chinese1300Not measuredOnly Gemini 2.5 Pro measured
LMArena Vision: reading text1270Not measuredOnly Gemini 2.5 Pro measured
LMArena Vision: humor1262Not measuredOnly Gemini 2.5 Pro measured
LMArena Vision: captioning1303Not measuredOnly Gemini 2.5 Pro measured
LMArena Vision: recognizing things1283Not measuredOnly Gemini 2.5 Pro measured
LMArena Vision: creative writing1265Not measuredOnly Gemini 2.5 Pro measured
LMArena Document1423Not measuredOnly Gemini 2.5 Pro measured
MultiChallenge53.6%Not measuredOnly Gemini 2.5 Pro measured
Neither model has a result on the other 13 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemini 2.5 Pro and Gemini 4 Neon
BenchmarkGemini 2.5 ProGemini 4 NeonLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Gemini 2.5 Pro$0.0063Fits in one request
Gemini 4 Neon$0.011Fits in one request

Gemini 2.5 Pro costs less

Repository review

50K input + 3K output tokens

Gemini 2.5 Pro$0.092Fits in one request
Gemini 4 Neon$0.195Fits in one request

Gemini 2.5 Pro costs less

Agent loop

200K input + 10K output tokens

Gemini 2.5 Pro$0.350Fits in one request
Gemini 4 Neon$0.750Fits in one request

Gemini 2.5 Pro costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Gemini 2.5 Pro and Gemini 4 Neon
SpecificationGoogle Gemini 2.5 ProGoogle Gemini 4 Neon
MakerGoogleGoogle
WeightsNo public weights listedNo public weights listed
ReleasedJune 17, 2025November 18, 2026
Context window1.05M1.05M
Input $/Mtok$1.25$3.00
Output $/Mtok$10.00$15.00
Coding index19.286.9
Independent measurements behind it34
Cost per solved task$0.64Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 2.5 Pro better than Gemini 4 Neon for coding?

Pick Gemini 4 Neon for coding: it scores 86.9 on the Vibe Coding Index against 19.2 for Gemini 2.5 Pro. Pick Gemini 2.5 Pro for output tokens 1.5x cheaper ($10.00 vs $15.00 per million).

Which is cheaper, Gemini 2.5 Pro or Gemini 4 Neon?

Gemini 2.5 Pro costs $1.25 per million input tokens and $10.00 per million output tokens; Gemini 4 Neon costs $3.00 and $15.00. Cost per solved task is known for Gemini 2.5 Pro only.

Which has the bigger context window, Gemini 2.5 Pro or Gemini 4 Neon?

Neither: both take 1.05M tokens.

How many benchmarks is the Gemini 2.5 Pro vs Gemini 4 Neon comparison based on?

38 benchmarks have a result for both models as of November 22, 2026. Gemini 2.5 Pro's coding index rests on 3 independent benchmarks and Gemini 4 Neon's on 4. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/gemini-2-5-pro-vs-gemini-4-neon" title="Gemini 2.5 Pro vs Gemini 4 Neon, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/gemini-2-5-pro-vs-gemini-4-neon">LLMBenchmark.ai, Gemini 2.5 Pro vs Gemini 4 Neon</a></p>