Share promptsSign inExplore benchmarks

Claude vs Gemini: AI benchmark scores compared (Oct 2026)

On the 20 public leaderboards where both have a reported score, Claude’s best-ranked model scores higher on 16, Gemini’s on 4. Counting only the 11 leaderboards their sources still update: Claude 10, Gemini 1.

In plain words

  • Multi-domain agent tasks: Claude higher on 1 of 1 (Vals Index).
  • Legal: Claude higher on all 4 (Legal Benchmarks: Contract Workflows, Legal Benchmarks: Data Extraction, LegalBench and Legal Research Bench).
  • Health & medical: Claude higher on 2 of 3 (HealthBench (independent run) and MedCode); Gemini higher on 1 of 3 (MedQA).
  • Finance: Claude higher on all 2 (Finance Agent and CorpFin v2).
  • Coding: Claude higher on all 2 (SWE-bench Verified and LiveCodeBench).
  • Agents & tool use: Claude higher on all 2 (Terminal-Bench 4.0 and Terminal-Bench 2.1).
  • Reasoning & knowledge: Claude higher on 2 of 4 (Humanity's Last Exam and MMLU Pro); Gemini higher on 2 of 4 (GPQA Diamond and AIME).
  • Chat preference: Gemini higher on 1 of 1 (LMArena Text Arena).
  • Multilingual: Claude higher on 1 of 1 (MGSM).
  • Claude’s best-ranked entry is most often Claude Opus 5.5 (5 of 20); Gemini’s is most often Gemini 4 Argon (5 of 20).

Claude here means Anthropic’s Claude models; Gemini means Google’s Gemini models. On each leaderboard we compare each side’s best-ranked model, so the model can differ by benchmark. A side only counts where it appears on the leaderboard (we keep each source’s top rows).

Results from published AI leaderboards. Compare models within each benchmark; scores from different tests aren’t directly comparable. How we collect these results

Multi-domain agent tasks

BenchmarkClaudeGeminiHigher score
Vals IndexAccuracy · dated 21 Oct 202670.85%Claude Sonnet 5.5 · #169.24%Gemini 4 Argon · #3Claude higherby 1.61 pts

Health & medical

BenchmarkClaudeGeminiHigher score
HealthBench (independent run)HealthBench · checked 22 Oct 202662.1%Claude Sonnet 5.5 (max with fallback) · #149.1%Gemini 4 Argon (high) · #13Claude higherby 13.0 pts
MedCodeAccuracy · dated 13 Oct 202663.57%Claude Opus 5 · #159.06%Gemini 3.1 Pro Preview (02/26) · #2Claude higherby 4.51 pts
MedQAAccuracy · dated 16 Apr 2026 · no longer updated95.88%Claude Opus 4.5 (Thinking) · #1096.37%Gemini 3.1 Pro Preview (02/26) · #3Gemini higherby 0.49 pts

Finance

BenchmarkClaudeGeminiHigher score
Finance AgentAccuracy · dated 19 Oct 202667.85%Claude Opus 5.5 · #167.66%Gemini 4 Argon · #2Claude higherby 0.19 pts
CorpFin v2Accuracy · dated 12 Aug 2026 · no longer updated73.19%Claude Opus 5 · #166.43%Gemini 3 Flash (12/25) · #20Claude higherby 6.76 pts

Coding

BenchmarkClaudeGeminiHigher score
SWE-bench VerifiedScore · dated 1 Sep 2026 · no longer updated97.0%Claude Opus 5 · #180.8%Gemini 3.7 Flash · #23Claude higherby 16.2 pts
LiveCodeBenchAccuracy · dated 1 Sep 2026 · no longer updated90.52%Claude Fable 5.1 · #189.48%Gemini 3.8 Flash · #3Claude higherby 1.04 pts

Agents & tool use

BenchmarkClaudeGeminiHigher score
Terminal-Bench 4.0Score · dated 19 Oct 202665.15%Claude Opus 5.5 · #157.58%Gemini 4 Argon · #6Claude higherby 7.57 pts
Terminal-Bench 2.1Score · dated 28 Sep 2026 · no longer updated87.64%Claude Opus 5.5 · #181.27%Gemini 3.8 Flash · #8Claude higherby 6.37 pts

Reasoning & knowledge

BenchmarkClaudeGeminiHigher score
Humanity's Last ExamAccuracy · checked 22 Oct 202646.50% ±2Fable 5.1 (xhigh) · #246.44% ±1.96gemini-3.1-pro-preview (thinking high) · #3Claude higherby 0.06 pts, within ±
GPQA DiamondAccuracy · dated 1 Sep 2026 · no longer updated93.43%Claude Opus 5 · #895.45%Gemini 3.1 Pro Preview (02/26) · #1Gemini higherby 2.02 pts
AIMEAccuracy · dated 16 Apr 2026 · no longer updated96.25%Claude Opus 4.7 · #798.13%Gemini 3.1 Pro Preview (02/26) · #1Gemini higherby 1.88 pts
MMLU ProAccuracy · dated 1 Sep 2026 · no longer updated92.38%Claude Fable 5.1 · #190.99%Gemini 3.1 Pro Preview (02/26) · #4Claude higherby 1.39 pts

Chat preference

BenchmarkClaudeGeminiHigher score
LMArena Text ArenaArena score · dated 20 Oct 20261509 ±7Claude Opus 5.5 (High) · #31524 ±9Gemini 4 Argon (High) · #1 · preliminaryGemini higherby 15, within ±

Multilingual

BenchmarkClaudeGeminiHigher score
MGSMAccuracy · dated 9 Jan 2026 · no longer updated95.20%Claude Opus 4.5 (Thinking) · #193.93%Gemini 3 Pro (11/25) · #7Claude higherby 1.27 pts

Every score links to the leaderboard that published it. Sources: Vals AI ↗ · Legal Benchmarks ↗ · Artificial Analysis ↗ · Scale AI and Center for AI Safety ↗ · LMArena ↗. ChatJudge does not run these tests.

Test Claude and Gemini on your own work

Run the same prompt in both and compare the answers. Prompts for Claude → · Prompts for Gemini →

ReviewsClaudeCompare two documents clause by clauseChecking a filled-in contract or proposal against its template.View prompt ProposalsChatGPTClarify commercial termsPricing, commission or partnership fees are being misread.View prompt
ResearchAny modelTurn messy text into a tableYou extract structured data from messy text. I will give you text and the fields I want. Return…View prompt AI Native SolutionsAI Native Solutions @ainativesolReport or remove
MarketingClaudeCustomer interviews to evidence-backed landing-page outlineFirst, extract exact quotes about desired outcomes, buying objections, and evidence customers ask…View prompt Nastation | AI for Client Work @NastaitonReport or remove
ResearchGeminiInvestor/stock due-diligenceAct as a forensic equity analyst. Conduct a deep research analysis for [COMPANY / TICKER] based on…View prompt MasterYoda @MasterYoda1991Report or remove
PricingClaudeUnit economics checkTesting whether your pricing survives your real usage costs.View prompt

Questions people ask

Is Claude better than Gemini?

On the 20 public leaderboards where both have a reported score, Claude’s best-ranked model scores higher on 16, Gemini’s on 4. Counting only the 11 leaderboards their sources still update: Claude 10, Gemini 1. Which is better depends on the task: the leaders differ by benchmark, and scores from different tests are not directly comparable.

Is Claude or Gemini better for multi-domain agent tasks?

Vals Index: Claude 70.85% vs Gemini 69.24%. Each figure is the best-ranked model from each side on that leaderboard, as published by its source.

Is Claude or Gemini better for legal work?

Legal Benchmarks: Contract Workflows: Claude 48.4% vs Gemini 25.8%; Legal Benchmarks: Data Extraction: Claude 76.7% vs Gemini 53.3%; LegalBench: Claude 88.56% vs Gemini 88.30%; Legal Research Bench: Claude 55.29% vs Gemini 54.81%. Each figure is the best-ranked model from each side on that leaderboard, as published by its source.

Is Claude or Gemini better for health & medical questions?

HealthBench (independent run): Claude 62.1% vs Gemini 49.1%; MedCode: Claude 63.57% vs Gemini 59.06%; MedQA: Gemini 96.37% vs Claude 95.88%. Each figure is the best-ranked model from each side on that leaderboard, as published by its source.

Is Claude or Gemini better for finance?

Finance Agent: Claude 67.85% vs Gemini 67.66%; CorpFin v2: Claude 73.19% vs Gemini 66.43%. Each figure is the best-ranked model from each side on that leaderboard, as published by its source.

Is Claude or Gemini better for coding?

SWE-bench Verified: Claude 97.0% vs Gemini 80.8%; LiveCodeBench: Claude 90.52% vs Gemini 89.48%. Each figure is the best-ranked model from each side on that leaderboard, as published by its source.

How recent is this comparison?

It is rebuilt from our benchmark data, last checked on 22 Oct 2026. Each leaderboard’s own date is shown in the table. MedQA, CorpFin v2, SWE-bench Verified, LiveCodeBench, Terminal-Bench 2.1, GPQA Diamond, AIME, MMLU Pro and MGSM are no longer updated by the source, so newer models may be missing there.

Which models are being compared?

On each leaderboard we take the best-ranked Claude model and the best-ranked Gemini model, so the model can change from one benchmark to the next. Claude’s top entry is most often Claude Opus 5.5 (5 of 20); Gemini’s is most often Gemini 4 Argon (5 of 20). Claude here means Anthropic’s Claude models; Gemini means Google’s Gemini models.