Claude Opus 4.6 vs Kimi K3
Wins 0 of 8 areas
—
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
Kimi K3 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 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 software04Kimi K34 of 4 tests
- AgentsCarrying out multi-step tasks on its own14Kimi K34 of 5 tests
- ReasoningHard problems that need careful thinking24Kimi K34 of 6 tests
- FactsGetting facts right instead of making them up02Kimi K32 of 3 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video02Kimi K32 of 2 tests
- MathCompetition and research-level math12Kimi K32 of 5 tests · 2 ties
- Long documentsFinding answers in very long texts01Kimi K31 of 1 test
- Following instructionsDoing exactly what it is asked01Kimi K31 of 1 test
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.
Kimi K3 costs 40% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Kimi K3 pulls ahead
- Multi-step tasks using many real software toolsMCP Atlas+21.5points ahead
- Reasoning about charts from research papersCharXiv (RQ)+13.9points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+13.2points ahead
Where Claude Opus 4.6 pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+8.4points ahead
- Common-sense trick questionsSimpleBench+6.9points ahead
- Fresh competition math problemsLiveBench · Mathematics+4.9points 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.
CodingKimi K3
- LiveBench · Agentic CodingKimi K3 by 13.24962.2+13.2
- LMArena · WebDevKimi K3 by 109 rating points15461655+109 rating
- SciCodeKimi K3 by 7.651.959.5+7.6
- LiveBench · CodingKimi K3 by 3.378.281.5+3.3
AgentsKimi K3
- MCP AtlasKimi K3 by 21.562.784.2+21.5
- OSWorld-VerifiedKimi K3 by 12.172.784.8+12.1
- AA ApexAgentsKimi K3 by 8.33341.3+8.3
- BrowseCompKimi K3 by 7.28491.2+7.2
- GDPValClaude Opus 4.6 by 4.255.951.7+4.2
ReasoningKimi K3
- CritPtKimi K3 by 10.812.623.4+10.8
- ARC-AGI-2Claude Opus 4.6 by 8.468.860.4+8.4
- Humanity's Last ExamKimi K3 by 739.946.9+7
- SimpleBenchClaude Opus 4.6 by 6.967.660.7+6.9
- GPQA DiamondKimi K3 by 3.989.693.5+3.9
- LiveBench · ReasoningKimi K3 by 288.790.7+2
FactsKimi K3
- AA-Omniscience · Non-hallucinationKimi K3 by 9.637.246.8+9.6
- SimpleQA VerifiedKimi K3 by 3.64750.6+3.6
- AA-Omniscience · Accuracytie4747.6tie
Images and chartsKimi K3
- CharXiv (RQ)Kimi K3 by 13.977.491.3+13.9
- MMMU-ProKimi K3 by 5.175.480.5+5.1
MathKimi K3
- FrontierMath Tier 4Kimi K3 by 12.226.839+12.2
- FrontierMath Tiers 1-3 (v2)Kimi K3 by 6.26672.2+6.2
- LiveBench · MathematicsClaude Opus 4.6 by 4.989.384.4+4.9
- HMMT Feb. 2026tie96.297tie
- AIME 2026tie96.796.7tie
Following instructionsKimi K3
- LiveBench · Instruction FollowingKimi K3 by 8.163.371.4+8.1
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.
- BabyVisionKimi K3 by 70.914.885.7+70.9
- MATH-VisionKimi K3 by 32.365.597.8+32.3
- Artificial Analysis Coding IndexKimi K3 by 28.148.176.2+28.1
- matharena_visual_math_overallKimi K3 by 1972.391.3+19
- AA Agentic IndexClaude Opus 4.6 by 1767.650.6+17
- ToolathlonKimi K3 by 16.456.873.2+16.4
- JobBenchKimi K3 by 16.336.652.9+16.3
- AA IntelligenceKimi K3 by 11.731.943.6+11.7
- livebench_data_analysisKimi K3 by 8.869.978.7+8.8
- NL2RepoKimi K3 by 8.249.858+8.2
- HLE (with tools)Kimi K3 by 6.553.359.8+6.5
- CyberGymKimi K3 by 6.273.880+6.2
- OfficeQA ProKimi K3 by 6.257.163.3+6.2
- AA-OmniscienceKimi K3 by 613.719.7+6
- DeepSearchQA (F1)Kimi K3 by 3.791.395+3.7
- livebench_languageKimi K3 by 2.283.385.5+2.2
- ARC-AGI-1Kimi K3 by 1.59394.5+1.5
Questions people ask
Which is better, Claude Opus 4.6 or Kimi K3?
Kimi K3 wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. Claude Opus 4.6 wins none.
Which is better for coding?
Kimi K3. It wins 4 of the 4 coding tests both models report; Claude Opus 4.6 wins none.
Which is cheaper?
Claude Opus 4.6 costs $5.00 per million input tokens and $25.00 per million output tokens; Kimi K3 costs $3.00 and $15.00. That makes Kimi K3 about 40% cheaper for the same work.
How do you compare the two?
We use the 44 benchmark tests both models have published scores on. The verdict counts the 27 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.