Claude Opus 4.8 vs Grok4.5
Wins 6 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math
Wins 1 of 8 areas
Long documents
Claude Opus 4.8 is the stronger all-rounder.Grok4.5 is cheaper and better at long documents.
Scores updated · 38 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.
- ReasoningHard problems that need careful thinking52Claude Opus 4.85 of 7 tests
- MathCompetition and research-level math30Claude Opus 4.83 of 3 tests
- CodingWriting and fixing software31Claude Opus 4.83 of 6 tests · 2 ties
- AgentsCarrying out multi-step tasks on its own21Claude Opus 4.82 of 3 tests
- FactsGetting facts right instead of making them up21Claude Opus 4.82 of 3 tests
- Images and chartsUnderstanding pictures, charts and video10Claude Opus 4.81 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts01Grok4.51 of 1 test
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
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.
Grok4.5 costs 73% 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
- Extremely hard research-level math problemsFrontierMath Tier 4+31.7points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+22.8points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+19.5points ahead
Where Grok4.5 pulls ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+7.9points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+6points ahead
- Common-sense trick questionsSimpleBench+5.2points ahead
Every test, side by side
All 38 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 4.8
- LiveBench · CodingClaude Opus 4.8 by 13.281.868.6+13.2
- LiveBench · Agentic CodingGrok4.5 by 650.556.5+6
- SWE-bench ProClaude Opus 4.8 by 4.569.264.7+4.5
- Terminal-Bench 2.1Claude Opus 4.8 by 384.681.6+3
- SciCodetie54.455tie
- LMArena · WebDevtie15561553tie
AgentsClaude Opus 4.8
- Terminal-Bench 4.0Claude Opus 4.8 by 11.121.710.6+11.1
- τ-Bench V3 · BankingGrok4.5 by 7.934.242.1+7.9
- GDPValClaude Opus 4.8 by 3.347.844.5+3.3
ReasoningClaude Opus 4.8
- ARC-AGI-2Claude Opus 4.8 by 19.572.152.6+19.5
- Humanity's Last ExamClaude Opus 4.8 by 648.742.7+6
- CritPtClaude Opus 4.8 by 5.520.915.4+5.5
- SimpleBenchGrok4.5 by 5.264.870+5.2
- LiveBench · ReasoningClaude Opus 4.8 by 289.287.2+2
- ARC-AGI-3Claude Opus 4.8 by 1.21.50.3+1.2
- GPQA DiamondGrok4.5 by 1.19293.1+1.1
FactsClaude Opus 4.8
- AA-Omniscience · Non-hallucinationClaude Opus 4.8 by 14.860.745.9+14.8
- SimpleQA VerifiedClaude Opus 4.8 by 4.75348.3+4.7
- AA-Omniscience · AccuracyGrok4.5 by 2.848.851.5+2.8
Images and chartsClaude Opus 4.8
- CharXiv (RQ)Claude Opus 4.8 by 8.389.981.6+8.3
- LMArena · Visiontie12951288tie
MathClaude Opus 4.8
- FrontierMath Tier 4Claude Opus 4.8 by 31.756.124.4+31.7
- FrontierMath Tiers 1-3 (v2)Claude Opus 4.8 by 22.88057.2+22.8
- LiveBench · MathematicsClaude Opus 4.8 by 3.594.390.8+3.5
Following instructionsEven
- LiveBench · Instruction Followingtie7271.5tie
Other results12 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.
- Terminal-Bench 4.0Claude Opus 4.8 by 11.223.612.4+11.2
- livebench_data_analysisGrok4.5 by 76673+7
- ARC-AGI-1Claude Opus 4.8 by 6.892.585.7+6.8
- GDPval-AA v2 EloClaude Opus 4.8 by 58 rating points15931535+58 rating
- DeepSWEClaude Opus 4.8 by 55853+5
- DeepSWE 1.1Claude Opus 4.8 by 55954+5
- AA-OmniscienceClaude Opus 4.8 by 3.528.825.3+3.5
- livebench_languageGrok4.5 by 3.179.782.8+3.1
- AA IntelligenceClaude Opus 4.8 by 341.838.8+3
- SWE-MarathonGrok4.5 by 32629+3
- Artificial Analysis Coding IndexClaude Opus 4.8 by 1.974.372.4+1.9
- AA Agentic Indextie42.642.1tie
Questions people ask
Which is better, Claude Opus 4.8 or Grok4.5?
Claude Opus 4.8 wins six of the eight areas we test: coding, agents, reasoning, facts, images and charts and math. Grok4.5 wins long documents, and costs 73% less. They are level on following instructions.
Which is better for coding?
Claude Opus 4.8. It wins 3 of the 6 coding tests both models report; Grok4.5 wins 1, and 2 are ties.
Which is cheaper?
Claude Opus 4.8 costs $5.00 per million input tokens and $25.00 per million output tokens; Grok4.5 costs $2.00 and $6.00. That makes Grok4.5 about 73% cheaper for the same work.
How do you compare the two?
We use the 38 benchmark tests both models have published scores on. The verdict counts the 26 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 12 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.