Claude 3.5 Sonnet vs Gemini 2.0 Flash Exp
Wins 0 of 3 areas
—
Wins 2 of 3 areas
Coding · Images and charts
Gemini 2.0 Flash Exp is the stronger all-rounder.
Scores updated · 44 tests both models report · How we compare
Where each one wins
Tests won in each of the three areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
Coding rests on a single test.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Gemini 2.0 Flash Exp pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+7.6points ahead
- Reads text in imagesOCRBench+5.6points ahead
- Math problems shown in pictures and chartsMathVista+5.4points ahead
Where Claude 3.5 Sonnet pulls ahead
- Common-sense trick questionsSimpleBench+8.6points ahead
Every test, side by side
All 44 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningEven
- SimpleBenchClaude 3.5 Sonnet by 8.627.518.9+8.6
- GPQA DiamondGemini 2.0 Flash Exp by 7.65663.6+7.6
- Humanity's Last Examtie3.24.1tie
Images and chartsGemini 2.0 Flash Exp
Other results36 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- Aider-PolyglotClaude 3.5 Sonnet by 23.145.322.2+23.1
- MATHGemini 2.0 Flash Exp by 1971.190.1+19
- OlympiadBenchGemini 2.0 Flash Exp by 17.728.446.1+17.7
- MMLUClaude 3.5 Sonnet by 1788.771.7+17
- AI2DClaude 3.5 Sonnet by 9.694.785.1+9.6
- MTOB eng → kalam (ChrF) no contextClaude 3.5 Sonnet by 820.212.2+8
- Chinese SimpleQA (C-SimpleQA)Gemini 2.0 Flash Exp by 7.955.463.3+7.9
- RULER 128KClaude 3.5 Sonnet by 7.893.886+7.8
- Arena HardClaude 3.5 Sonnet by 6.579.272.7+6.5
- naturalquestions_closedbookClaude 3.5 Sonnet by 5.950.244.3+5.9
- MTOB kalam → eng (BLEURT) full bookClaude 3.5 Sonnet by 5.362.357+5.3
- HumanEvalClaude 3.5 Sonnet by 4.193.789.6+4.1
- MTOB eng → kalam (ChrF) half bookClaude 3.5 Sonnet by 4.153.649.5+4.1
- NarrativeQAGemini 2.0 Flash Exp by 3.774.678.3+3.7
- OpenBookQAClaude 3.5 Sonnet by 2.697.294.6+2.6
- ChartQAClaude 3.5 Sonnet by 2.590.888.3+2.5
- ChartQA_relaxedClaude 3.5 Sonnet by 2.590.888.3+2.5
- MEGA-Bench_macroGemini 2.0 Flash Exp by 2.551.453.9+2.5
- MTOB eng → kalam (ChrF) full bookClaude 3.5 Sonnet by 2.455.653.3+2.4
- MTOB kalam → eng (BLEURT) no contextGemini 2.0 Flash Exp by 2.431.433.8+2.4
- DocVQAClaude 3.5 Sonnet by 2.395.292.9+2.3
- MTOB kalam → eng (BLEURT) half bookClaude 3.5 Sonnet by 2.259.757.5+2.2
- IFEvalGemini 2.0 Flash Exp by 1.986.588.4+1.9
- GSM8KClaude 3.5 Sonnet by 1.896.494.6+1.8
- SimpleQAClaude 3.5 Sonnet by 1.828.426.6+1.8
- IFEval (avg)Claude 3.5 Sonnet by 1.790.188.4+1.7
- RULER 64KClaude 3.5 Sonnet by 1.595.293.7+1.5
- MMLU-ProClaude 3.5 Sonnet by 1.277.676.4+1.2
- AA IntelligenceGemini 2.0 Flash Exp by 17.28.2+1
- MBPP+ (EvalPlus-augmented)tie75.175.9tie
- Ruler 32ktie9595.7tie
- Ruler 16ktie95.795.1tie
- DROP (F1)tie88.889.3tie
- Ruler 4ktie96.596tie
- M-LongDoc (multimodal long-document benchmark)tie31.431.4tie
- Ruler 8ktie9696tie
Questions people ask
Which is better, Claude 3.5 Sonnet or Gemini 2.0 Flash Exp?
Gemini 2.0 Flash Exp wins two of the three areas where both have results: coding and images and charts. Claude 3.5 Sonnet wins none. They are level on reasoning.
Which is better for coding?
Gemini 2.0 Flash Exp. It wins the one coding test both models report.
How do you compare the two?
We use the 44 benchmark tests both models have published scores on. The verdict counts the 8 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 36 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.