Claude Sonnet 4.6 vs Gemini 3 Pro
Wins 3 of 7 areas
Coding · Agents · Long documents
Wins 3 of 7 areas
Reasoning · Images and charts · Following instructions
The two are evenly matched.Claude Sonnet 4.6 is better at agents and coding; Gemini 3 Pro at images and charts and reasoning.
Scores updated · 40 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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.
- AgentsCarrying out multi-step tasks on its own40Claude Sonnet 4.64 of 4 tests
- CodingWriting and fixing software41Claude Sonnet 4.64 of 5 tests
- Long documentsFinding answers in very long texts20Claude Sonnet 4.62 of 2 tests
- Images and chartsUnderstanding pictures, charts and video03Gemini 3 Pro3 of 3 tests
- ReasoningHard problems that need careful thinking13Gemini 3 Pro3 of 4 tests
- Following instructionsDoing exactly what it is asked01Gemini 3 Pro1 of 1 test
- FactsGetting facts right instead of making them up22Even2 each
Following instructions 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.
Gemini 3 Pro costs 22% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Claude Sonnet 4.6 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+43.1points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+27.2points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+15.5points ahead
Where Gemini 3 Pro pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+40.1points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+14.9points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+13.8points ahead
Every test, side by side
All 40 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 4.6
- SWE-bench ProClaude Sonnet 4.6 by 14.858.143.3+14.8
- Terminal-Bench HardClaude Sonnet 4.6 by 11.35341.7+11.3
- LMArena · WebDevClaude Sonnet 4.6 by 82 rating points15221440+82 rating
- SciCodeGemini 3 Pro by 650.156.1+6
- SWE-bench VerifiedClaude Sonnet 4.6 by 3.479.676.2+3.4
AgentsClaude Sonnet 4.6
- BrowseCompClaude Sonnet 4.6 by 15.574.759.2+15.5
- AA ApexAgentsClaude Sonnet 4.6 by 9.62818.4+9.6
- MCP AtlasClaude Sonnet 4.6 by 7.261.354.1+7.2
- GDPValClaude Sonnet 4.6 by 2.536.734.2+2.5
ReasoningGemini 3 Pro
- ARC-AGI-2Claude Sonnet 4.6 by 27.258.331.1+27.2
- Humanity's Last ExamGemini 3 Pro by 6.133.639.7+6.1
- CritPtGemini 3 Pro by 63.19.1+6
- GPQA DiamondGemini 3 Pro by 3.387.590.8+3.3
FactsEven
- AA-Omniscience · Non-hallucinationClaude Sonnet 4.6 by 43.151.68.5+43.1
- SimpleQA VerifiedGemini 3 Pro by 40.132.872.9+40.1
- AA-Omniscience · AccuracyGemini 3 Pro by 14.940.955.8+14.9
- Vectara HHEM hallucination ratelower is betterClaude Sonnet 4.6 by 310.613.6+3
Images and chartsGemini 3 Pro
- CharXiv (RQ)Gemini 3 Pro by 9.871.681.4+9.8
- MMMU-ProGemini 3 Pro by 6.973.380.2+6.9
- LMArena · VisionGemini 3 Pro by 22 rating points12831305+22 rating
Long documentsClaude Sonnet 4.6
- MRCR v2 8 needle 128k (average)Claude Sonnet 4.6 by 7.984.977+7.9
- AA-LCRClaude Sonnet 4.6 by 48076+4
Following instructionsGemini 3 Pro
- IFBenchGemini 3 Pro by 13.856.670.4+13.8
Other results17 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.
- GDPval-AA (Elo)Claude Sonnet 4.6 by 481 rating points16761195+481 rating
- AA Agentic IndexGemini 3 Pro by 18.933.152+18.9
- Artificial Analysis Coding IndexClaude Sonnet 4.6 by 16.56346.5+16.5
- τ²-Bench Telecom (AA run)Gemini 3 Pro by 11.475.787.1+11.4
- ARC-AGI-1Claude Sonnet 4.6 by 118675+11
- frontiermath_tier_4_v1Gemini 3 Pro by 10.58.318.8+10.5
- τ²-Bench (Retail)Claude Sonnet 4.6 by 6.491.785.3+6.4
- Terminal-Bench 2.0Claude Sonnet 4.6 by 4.959.154.2+4.9
- ToolathlonClaude Sonnet 4.6 by 4.64136.4+4.6
- AA-OmniscienceGemini 3 Pro by 3.112.215.3+3.1
- vectara_factual_consistencyClaude Sonnet 4.6 by 389.486.4+3
- MMMLUGemini 3 Pro by 2.589.391.8+2.5
- AA IntelligenceClaude Sonnet 4.6 by 2.130.128+2.1
- LiveBenchClaude Sonnet 4.6 by 2.175.573.4+2.1
- vectara_avg_summary_lengthClaude Sonnet 4.6 by 12.8 rating points114.7101.9+12.8 rating
- HLE (with tools)Claude Sonnet 4.6 by 146.845.8+1
- vectara_answer_ratetie99.999.4tie
Questions people ask
Which is better, Claude Sonnet 4.6 or Gemini 3 Pro?
Claude Sonnet 4.6 and Gemini 3 Pro each win three of the seven areas where both have results. Claude Sonnet 4.6 wins coding, agents and long documents; Gemini 3 Pro wins reasoning, images and charts and following instructions. Gemini 3 Pro costs 22% less. They are level on facts.
Which is better for coding?
Claude Sonnet 4.6. It wins 4 of the 5 coding tests both models report; Gemini 3 Pro wins 1.
Which is cheaper?
Claude Sonnet 4.6 costs $3.00 per million input tokens and $15.00 per million output tokens; Gemini 3 Pro costs $2.00 and $12.00. That makes Gemini 3 Pro about 22% cheaper for the same work.
How do you compare the two?
We use the 40 benchmark tests both models have published scores on. The verdict counts the 23 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 17 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.