Skip to main content

Model profile

Gemini 2.5 Pro

GoogleEstablishedReleased March 2025
Data verified
Overall Score
57.25Public #70 of 200Verified #45 of 99
Arena Elo
1446
Eligible category ranks
2of 8
Price (1M tokens)
$1.25 in / $10 out
API pricing
Speed
117tok/s
Context
1M

Evidence coverage

24 of 321 tracked benchmarks are published. 6 are verified and 18 provisional. 7 of 8 categories are measured.

Updated July 20, 2026Methodology
Published / tracked
24 / 321
Verified
6
Provisional
18
Categories with evidence
7 / 8

Evidence by category

  • Agentic5 benchmarks
    Mixed evidence
  • Coding4 benchmarks
    Mixed evidence
  • Reasoning2 benchmarks
    Reported
  • Knowledge8 benchmarks
    Mixed evidence
  • Math2 benchmarks
    Verified
  • Multilingual0 benchmarks
    Not measured
  • Multimodal2 benchmarks
    Reported
  • Inst. Following1 benchmark
    Reported
ProprietaryNon-Reasoning
Confidence:
Low
base

Gemini 2.5 Pro ranks #70 out of 200 models on the public leaderboard with an overall score of 57.25/100. It also ranks #45 out of 99 on the verified leaderboard. While not a frontier model, it offers specific advantages depending on the use case.

Gemini 2.5 Pro is a proprietary model with a 1M token context window. It processes queries without explicit chain-of-thought reasoning, offering faster response times and lower token usage.

This profile currently has 24 of 321 tracked benchmarks. BenchLM only exposes non-generated benchmark rows publicly, so missing categories stay blank until a sourced evaluation is available.

Its strongest category is Agentic (#60), while its weakest is Coding (#111). This performance profile makes it particularly useful for coding agents, browser research, and computer-use workflows.

Peer position

Exact provisional scores and ranks for the closest listed peers. A score can appear before a model clears the evidence threshold for a rank, so equal scores can have different rank states.

Range 56.7957.56

  1. GLM-4.5
    Z.AI
    #6857.56
    GLM-4.5 is #68 with a score of 57.56.
    Compare
  2. Claude Opus 4.5 Thinking
    Anthropic
    #6957.44
    Claude Opus 4.5 Thinking is #69 with a score of 57.44.
    Compare
  3. Gemini 2.5 ProCurrent model
    Google
    #7057.25
    Gemini 2.5 Pro is #70 with a score of 57.25.
  4. Qwen3.5 397B
    Alibaba
    #7157.01
    Qwen3.5 397B is #71 with a score of 57.01.
    Compare
  5. Qwen3.5-35B-A3B
    Alibaba
    #7256.97
    Qwen3.5-35B-A3B is #72 with a score of 56.97.
    Compare
  6. GPT-5.3-Codex-Spark
    OpenAI
    #7356.91
    GPT-5.3-Codex-Spark is #73 with a score of 56.91.
    Compare
  7. Kimi K2.6
    Moonshot AI
    #7456.79
    Kimi K2.6 is #74 with a score of 56.79.
    Compare

Category percentile

More

Relative position among models eligible for each sourced category. A higher percentile means a stronger position within that category's ranked cohort; 100 is highest.

  1. Agentic50%
    Eligible cohort rank #60 of 119Category score 48.1
  2. Coding9%
    Eligible cohort rank #111 of 122Category score 35.7

Category evidence

Scores and ranks appear only where this model has published benchmark evidence. Categories without displayable source records remain not measured.

Category scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticRank #60 of 119Percentile 50thWeight 22%5 benchmarksMixed sources48.1
CodingRank #111 of 122Percentile 9thWeight 20%4 benchmarksMixed sources35.7
ReasoningRank Not rankedWeight 17%2 benchmarksReported57.6
KnowledgeRank Not rankedWeight 12%8 benchmarksMixed sources23.8
MathRank Not rankedWeight 5%2 benchmarksVerified35.5
MultilingualWeight 7%0 benchmarksNot measuredNot measured
MultimodalRank Not rankedWeight 12%2 benchmarksReported85.2
Inst. FollowingRank Not rankedWeight 5%1 benchmarkReported83.0

Chatbot Arena performance

Scroll horizontally to inspect confidence intervals and vote counts.

Chatbot Arena Elo, confidence interval, and vote count by evaluation view
ViewEloConfidence intervalVotes
Text Overall1446±2.5124,422
Coding1465±4.326,537
Math1441±7.07,635
Instruction Following1439±3.934,227
Creative Writing1444±5.217,398
Multi-turn1448±4.821,182
Hard Prompts1458±3.264,580
Hard Prompts (English)1457±4.032,429
Longer Query1456±4.132,550

Benchmark Details

Rows below have a displayable published verification record. Each source link and provenance note remains in the page HTML while its category is closed. Source-unverified manual rows and generated rows stay hidden.

Agentic5 benchmarks
AA Agentic IndexReported

Artificial Analysis Agentic Index

7.1%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
τ²-bench resultsReported

τ²-Bench Tool-Agent-User Evaluation

54.1%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Gert LabsBenchmark exact

Gert Labs Composite Game Benchmark

42.01%Display only
Source: Gert Labs rankingsProvenance: Gert Labs reports this composite leaderboard score in the public rankings API. BenchLM scales the source gscore from 0-1 to 0-100 and stores it as a display-only agentic benchmark.
GDPval-AAReported

GDPval-AA normalized

8.3%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
GDPval-AAReported
665Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Coding4 benchmarks
SWE-bench VerifiedProvider exact

Software Engineering Benchmark Verified

63.8%Weighted 16%
Source: Google: Gemini 2.5 ProProvenance: Provider exact
Vibe Code BenchBenchmark exact

Vibe Code Bench v1.1

0.40%Display only
Source: Vals AI: Vibe Code Bench v1.1Provenance: Vals Vibe Code Bench v1.1 reports this exact row under google/gemini-2.5-pro; BenchLM stores it on the local vibeCodeBench key.
AA Coding IndexReported

Artificial Analysis Coding Index

33.3%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-SciCodeReported

Artificial Analysis SciCode

42.8%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Reasoning2 benchmarks
AA-LCRReported

Artificial Analysis Long Context Reasoning

66.0%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
CritPtReported

Critical Physics Tasks

2.6%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Knowledge8 benchmarks
HLEProvider exact

Humanity's Last Exam

18.8%Weighted 45%
Source: Google: Gemini 2.5 ProProvenance: Provider exact
GPQAReported

Graduate-Level Google-Proof Q&A

83%Weighted 7%
Source: Reported upstream sourceProvenance: Reported row carried from an upstream public source. Displayable on BenchLM, but not treated as verified unless explicitly marked otherwise.
Artificial Analysis Intelligence IndexReported
25.8%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-GPQA DiamondReported

Artificial Analysis GPQA Diamond

84.4%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-HLEReported

Artificial Analysis Humanity's Last Exam

21.1%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience IndexReported

Artificial Analysis Omniscience Index

-14.3%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience AccuracyReported

Artificial Analysis Omniscience Accuracy

39.0%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience Hallucination RateReported

Artificial Analysis Omniscience Hallucination Rate

87.4%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Math2 benchmarks
FrontierMath v2 (Tiers 1-3)Benchmark exact

FrontierMath v2 Tiers 1-3

14.138%Weighted 30%
Source: Epoch AI FrontierMath v2 leaderboardProvenance: Epoch AI reports FrontierMath v2 Tiers 1-3 at 14.138% for gemini-2.5-pro. BenchLM selects the highest published thinking effort for the model and stores the v2 benchmark slice separately.
FrontierMath v2 (Tier 4)Benchmark exact

FrontierMath v2 Tier 4

4.167%Weighted 10%
Source: Epoch AI FrontierMath v2 leaderboardProvenance: Epoch AI reports FrontierMath v2 Tier 4 at 4.167% for gemini-2.5-pro. BenchLM selects the highest published thinking effort for the model and stores the v2 benchmark slice separately.
Multimodal2 benchmarks
AA-MMMU-ProReported

Artificial Analysis MMMU-Pro

74.9%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Design Arena WebsiteReported

Design Arena Website Elo

1199Display only
Source: OpenRouter model benchmarksProvenance: Display-only Design Arena Website Elo synced from OpenRouter model benchmark metadata. It is excluded from BenchLM weighted scoring.
Inst. Following1 benchmark
AA-IFBenchReported

Artificial Analysis IFBench

48.7%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.

Frequently Asked Questions

How does Gemini 2.5 Pro perform overall in AI benchmarks?

Gemini 2.5 Pro has 24 published benchmark scores on BenchLM, but it does not yet have enough non-generated coverage to receive a global overall rank.

Is Gemini 2.5 Pro good for knowledge and understanding?

Gemini 2.5 Pro has visible benchmark coverage in knowledge and understanding, but BenchLM does not currently assign it a global category rank there.

Is Gemini 2.5 Pro good for coding and programming?

Gemini 2.5 Pro ranks #111 out of 122 models in coding and programming benchmarks with an average score of 35.7. There are stronger options in this category.

Is Gemini 2.5 Pro good for mathematics?

Gemini 2.5 Pro has visible benchmark coverage in mathematics, but BenchLM does not currently assign it a global category rank there.

Is Gemini 2.5 Pro good for reasoning and logic?

Gemini 2.5 Pro has visible benchmark coverage in reasoning and logic, but BenchLM does not currently assign it a global category rank there.

Is Gemini 2.5 Pro good for agentic tool use and computer tasks?

Gemini 2.5 Pro ranks #60 out of 119 models in agentic tool use and computer tasks benchmarks with an average score of 48.1. There are stronger options in this category.

Is Gemini 2.5 Pro good for multimodal and grounded tasks?

Gemini 2.5 Pro has visible benchmark coverage in multimodal and grounded tasks, but BenchLM does not currently assign it a global category rank there.

Is Gemini 2.5 Pro good for instruction following?

Gemini 2.5 Pro has visible benchmark coverage in instruction following, but BenchLM does not currently assign it a global category rank there.

Does Gemini 2.5 Pro have full benchmark coverage on BenchLM?

Not yet. Gemini 2.5 Pro currently has 24 published benchmark scores out of the 321 benchmarks BenchLM tracks. BenchLM only exposes non-generated public benchmark rows, so missing categories stay blank until a sourced evaluation is available.

What is the context window size of Gemini 2.5 Pro?

Gemini 2.5 Pro has a published context window of 1M, which determines how much text it can process in a single interaction.

Last updated: July 20, 2026 · Runtime metrics stay blank until BenchLM has a sourced snapshot.

Choose with this week’s evidence

Join 2,000+ readers for ranking moves, new releases, pricing changes, and the evidence behind them.

Free. One email per week.