Claude Opus 4.8 vs Gemini 3 Pro
Wins 5 of 8 areas
Coding · Agents · Reasoning · Math · Long documents
Wins 2 of 8 areas
Facts · Following instructions
Claude Opus 4.8 is the stronger all-rounder.Gemini 3 Pro is cheaper and better at facts.
Scores updated · 41 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- CodingWriting and fixing software51Claude Opus 4.85 of 6 tests
- AgentsCarrying out multi-step tasks on its own40Claude Opus 4.84 of 4 tests
- ReasoningHard problems that need careful thinking41Claude Opus 4.84 of 5 tests
- MathCompetition and research-level math20Claude Opus 4.82 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts10Claude Opus 4.81 of 1 test
- FactsGetting facts right instead of making them up12Gemini 3 Pro2 of 3 tests
- Following instructionsDoing exactly what it is asked01Gemini 3 Pro1 of 1 test
- Images and chartsUnderstanding pictures, charts and video11Even1 each
Long documents and following instructions rest on a single test each.
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 53% 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 Opus 4.8 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+52.2points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+41points ahead
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+25.9points ahead
Where Gemini 3 Pro pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+19.9points ahead
- Common-sense trick questionsSimpleBench+11.6points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+8.2points ahead
Every test, side by side
All 41 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 4.8
- SWE-bench ProClaude Opus 4.8 by 25.969.243.3+25.9
- SWE-bench MultilingualClaude Opus 4.8 by 19.484.465+19.4
- Terminal-Bench HardClaude Opus 4.8 by 16.658.341.7+16.6
- SWE-bench VerifiedClaude Opus 4.8 by 12.488.676.2+12.4
- LMArena · WebDevClaude Opus 4.8 by 116 rating points15561440+116 rating
- SciCodeGemini 3 Pro by 1.754.456.1+1.7
AgentsClaude Opus 4.8
- BrowseCompClaude Opus 4.8 by 25.184.359.2+25.1
- MCP AtlasClaude Opus 4.8 by 23.777.854.1+23.7
- AA ApexAgentsClaude Opus 4.8 by 2139.418.4+21
- GDPValClaude Opus 4.8 by 13.647.834.2+13.6
ReasoningClaude Opus 4.8
- ARC-AGI-2Claude Opus 4.8 by 4172.131.1+41
- CritPtClaude Opus 4.8 by 11.820.99.1+11.8
- SimpleBenchGemini 3 Pro by 11.664.876.4+11.6
- Humanity's Last ExamClaude Opus 4.8 by 948.739.7+9
- GPQA DiamondClaude Opus 4.8 by 1.29290.8+1.2
FactsGemini 3 Pro
- AA-Omniscience · Non-hallucinationClaude Opus 4.8 by 52.260.78.5+52.2
- SimpleQA VerifiedGemini 3 Pro by 19.95372.9+19.9
- AA-Omniscience · AccuracyGemini 3 Pro by 748.855.8+7
Images and chartsEven
- CharXiv (RQ)Claude Opus 4.8 by 8.589.981.4+8.5
- LMArena · VisionGemini 3 Pro by 10 rating points12951305+10 rating
MathClaude Opus 4.8
- HMMT Feb. 2026Claude Opus 4.8 by 10.396.786.4+10.3
- AIME 2026Claude Opus 4.8 by 495.791.7+4
- IMOAnswerBenchtie83.583.1tie
Following instructionsGemini 3 Pro
- IFBenchGemini 3 Pro by 8.262.270.4+8.2
Other results16 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 Opus 4.8 by 695 rating points18901195+695 rating
- DeepSearchQA (F1)Claude Opus 4.8 by 29.993.163.2+29.9
- Artificial Analysis Coding IndexClaude Opus 4.8 by 27.874.346.5+27.8
- ToolathlonClaude Opus 4.8 by 23.559.936.4+23.5
- Terminal-Bench 2.0Claude Opus 4.8 by 20.474.654.2+20.4
- ARC-AGI-1Claude Opus 4.8 by 17.592.575+17.5
- ScreenSpot-Pro (No tools)Claude Opus 4.8 by 15.287.972.7+15.2
- AA IntelligenceClaude Opus 4.8 by 13.841.828+13.8
- AA-OmniscienceClaude Opus 4.8 by 13.528.815.3+13.5
- frontiermath_tier_4_v1Claude Opus 4.8 by 12.531.318.8+12.5
- HLE (with tools)Claude Opus 4.8 by 12.157.945.8+12.1
- AA Agentic IndexGemini 3 Pro by 9.442.652+9.4
- τ²-Bench Telecom (AA run)Claude Opus 4.8 by 7.394.487.1+7.3
- LiveBenchClaude Opus 4.8 by 3.877.273.4+3.8
- HMMT Nov. 2025Claude Opus 4.8 by 3.296.593.3+3.2
- matharena_visual_math_overallGemini 3 Pro by 2.681.684.2+2.6
Questions people ask
Which is better, Claude Opus 4.8 or Gemini 3 Pro?
Claude Opus 4.8 wins five of the eight areas we test: coding, agents, reasoning, math and long documents. Gemini 3 Pro wins facts and following instructions, and costs 53% less. They are level on images and charts.
Which is better for coding?
Claude Opus 4.8. It wins 5 of the 6 coding tests both models report; Gemini 3 Pro wins 1.
Which is cheaper?
Claude Opus 4.8 costs $5.00 per million input tokens and $25.00 per million output tokens; Gemini 3 Pro costs $2.00 and $12.00. That makes Gemini 3 Pro about 53% cheaper for the same work.
How do you compare the two?
We use the 41 benchmark tests both models have published scores on. The verdict counts the 25 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 16 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.