Claude Opus 4.6 vs GPT-5.1
Wins 4 of 7 areas
Coding · Agents · Reasoning · Images and charts
Wins 3 of 7 areas
Facts · Long documents · Following instructions
Claude Opus 4.6 wins more areas, narrowly.GPT-5.1 is cheaper and better at following instructions.
Scores updated · 39 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.
- ReasoningHard problems that need careful thinking50Claude Opus 4.65 of 5 tests
- CodingWriting and fixing software40Claude Opus 4.64 of 5 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own20Claude Opus 4.62 of 2 tests
- Images and chartsUnderstanding pictures, charts and video10Claude Opus 4.61 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked02GPT-5.12 of 2 tests
- FactsGetting facts right instead of making them up12GPT-5.12 of 4 tests · 1 tie
- Long documentsFinding answers in very long texts01GPT-5.11 of 1 test
Long documents 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.
GPT-5.1 costs 63% 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 4.6 pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+51.2points ahead
- Real work tasks from 44 professionsGDPVal+39.4points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+33.2points ahead
Where GPT-5.1 pulls ahead
- Keeps track of context across a multi-turn chatMulti-Challenge+26.2points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+19.8points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+10.9points ahead
Every test, side by side
All 39 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 4.6
- LMArena · WebDevClaude Opus 4.6 by 151 rating points15461395+151 rating
- SciCodeClaude Opus 4.6 by 8.651.943.3+8.6
- SWE-bench VerifiedClaude Opus 4.6 by 4.580.876.3+4.5
- LiveCodeBench v6Claude Opus 4.6 by 1.888.887+1.8
- Terminal-Bench Hardtie46.245.5tie
AgentsClaude Opus 4.6
- GDPValClaude Opus 4.6 by 39.455.916.5+39.4
- BrowseCompClaude Opus 4.6 by 33.28450.8+33.2
ReasoningClaude Opus 4.6
- ARC-AGI-2Claude Opus 4.6 by 51.268.817.6+51.2
- SimpleBenchClaude Opus 4.6 by 14.467.653.2+14.4
- Humanity's Last ExamClaude Opus 4.6 by 11.439.928.5+11.4
- CritPtClaude Opus 4.6 by 7.712.64.9+7.7
- GPQA DiamondClaude Opus 4.6 by 2.389.687.3+2.3
FactsGPT-5.1
- AA-Omniscience · Non-hallucinationGPT-5.1 by 10.937.248.1+10.9
- AA-Omniscience · AccuracyClaude Opus 4.6 by 9.34737.7+9.3
- SimpleQA VerifiedGPT-5.1 by 14748+1
- Vectara HHEM hallucination ratelower is bettertie12.212.1tie
Images and chartsClaude Opus 4.6
- LMArena · VisionClaude Opus 4.6 by 66 rating points13161250+66 rating
- MMMU-Protie75.475.5tie
Following instructionsGPT-5.1
- Multi-ChallengeGPT-5.1 by 26.237.163.4+26.2
- IFBenchGPT-5.1 by 19.853.172.9+19.8
Other results18 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.
- AA Agentic IndexClaude Opus 4.6 by 4667.621.6+46
- ARC-AGI-1Claude Opus 4.6 by 20.29372.8+20.2
- Terminal-Bench 2.0Claude Opus 4.6 by 17.865.447.6+17.8
- τ²-Bench (Retail)Claude Opus 4.6 by 1491.977.9+14
- vectara_avg_summary_lengthGPT-5.1 by 116.8 rating points137.6254.4+116.8 rating
- HLE (with tools)Claude Opus 4.6 by 10.653.342.7+10.6
- frontiermath_tier_4_v1Claude Opus 4.6 by 10.422.912.5+10.4
- τ²-Bench Telecom (AA run)Claude Opus 4.6 by 10.292.181.9+10.2
- AA-OmniscienceClaude Opus 4.6 by 8.313.75.4+8.3
- AA IntelligenceClaude Opus 4.6 by 7.231.924.7+7.2
- AIME 2025Claude Opus 4.6 by 5.899.894+5.8
- HMMT Nov. 2025Claude Opus 4.6 by 4.696.391.7+4.6
- matharena_visual_math_overallGPT-5.1 by 4.672.376.9+4.6
- MMLU-ProClaude Opus 4.6 by 2.189.187+2.1
- Artificial Analysis Coding IndexGPT-5.1 by 1.348.149.4+1.3
- vectara_answer_ratetie99.8100tie
- MMMLUtie91.191tie
- vectara_factual_consistencytie87.887.9tie
Questions people ask
Which is better, Claude Opus 4.6 or GPT-5.1?
Claude Opus 4.6 wins four of the seven areas where both have results: coding, agents, reasoning and images and charts. GPT-5.1 wins facts, long documents and following instructions, and costs 63% less.
Which is better for coding?
Claude Opus 4.6. It wins 4 of the 5 coding tests both models report; GPT-5.1 wins none, and 1 is a tie.
Which is cheaper?
Claude Opus 4.6 costs $5.00 per million input tokens and $25.00 per million output tokens; GPT-5.1 costs $1.25 and $10.00. That makes GPT-5.1 about 63% cheaper for the same work.
How do you compare the two?
We use the 39 benchmark tests both models have published scores on. The verdict counts the 21 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.