Claude Opus 4.5 vs Kimi K2.6
Wins 1 of 8 areas
Facts
Wins 7 of 8 areas
Coding · Agents · Reasoning · Images and charts · Math · Long documents · Following instructions
Kimi K2.6 is the stronger all-rounder.Claude Opus 4.5 is better at facts.
Scores updated · 52 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 software26Kimi K2.66 of 9 tests · 1 tie
- MathCompetition and research-level math15Kimi K2.65 of 6 tests
- ReasoningHard problems that need careful thinking03Kimi K2.63 of 4 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own13Kimi K2.63 of 4 tests
- Images and chartsUnderstanding pictures, charts and video02Kimi K2.62 of 2 tests
- Following instructionsDoing exactly what it is asked02Kimi K2.62 of 2 tests
- Long documentsFinding answers in very long texts01Kimi K2.61 of 1 test
- FactsGetting facts right instead of making them up21Claude Opus 4.52 of 4 tests · 1 tie
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.
Kimi K2.6 costs 84% 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 K2.6 pulls ahead
- Fixes real GitHub issues in many programming languagesSWE-bench Multilingual+69.7points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+49.3points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+22.8points ahead
Where Claude Opus 4.5 pulls ahead
- Real work tasks from 44 professionsGDPVal+20.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+14points ahead
- Short factual questions, answered correctlySimpleQA Verified+10.8points ahead
Every test, side by side
All 52 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingKimi K2.6
- SWE-bench MultilingualKimi K2.6 by 69.7776.7+69.7
- LiveBench · Agentic CodingKimi K2.6 by 7.239.746.9+7.2
- SWE-bench ProKimi K2.6 by 6.65258.6+6.6
- LiveCodeBench v6Kimi K2.6 by 4.884.889.6+4.8
- LMArena · WebDevKimi K2.6 by 40 rating points14691509+40 rating
- Terminal-Bench HardClaude Opus 4.5 by 3.14743.9+3.1
- SciCodeKimi K2.6 by 249.551.5+2
- LiveBench · CodingClaude Opus 4.5 by 1.179.778.6+1.1
- SWE-bench Verifiedtie80.980.2tie
AgentsKimi K2.6
- BrowseCompKimi K2.6 by 49.33786.3+49.3
- GDPValClaude Opus 4.5 by 20.347.327+20.3
- OSWorld-VerifiedKimi K2.6 by 6.866.373.1+6.8
- MCP AtlasKimi K2.6 by 5.862.368.1+5.8
ReasoningKimi K2.6
- Humanity's Last ExamKimi K2.6 by 7.430.137.5+7.4
- GPQA DiamondKimi K2.6 by 4.586.691.1+4.5
- CritPtKimi K2.6 by 3.44.68+3.4
- LiveBench · Reasoningtie80.179.4tie
FactsClaude Opus 4.5
- AA-Omniscience · Non-hallucinationKimi K2.6 by 20.53959.5+20.5
- AA-Omniscience · AccuracyClaude Opus 4.5 by 1446.632.6+14
- SimpleQA VerifiedClaude Opus 4.5 by 10.845.734.9+10.8
- Vectara HHEM hallucination ratelower is bettertie10.910.8tie
Images and chartsKimi K2.6
- CharXiv (RQ)Kimi K2.6 by 19.567.286.7+19.5
- MMMU-ProKimi K2.6 by 5.47479.4+5.4
MathKimi K2.6
- FrontierMath Tiers 1-3 (v2)Kimi K2.6 by 22.834.457.2+22.8
- FrontierMath Tier 4Kimi K2.6 by 20.74.925.6+20.7
- HMMT Feb. 2026Kimi K2.6 by 7.485.392.7+7.4
- LiveBench · MathematicsClaude Opus 4.5 by 6.190.484.3+6.1
- IMOAnswerBenchKimi K2.6 by 28486+2
- AIME 2026Kimi K2.6 by 1.395.196.4+1.3
Following instructionsKimi K2.6
- IFBenchKimi K2.6 by 185876+18
- LiveBench · Instruction FollowingKimi K2.6 by 1.862.564.4+1.8
Other results20 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 4.5 by 37.559.622.1+37.5
- MATH-VisionKimi K2.6 by 16.177.193.2+16.1
- Artificial Analysis Coding IndexKimi K2.6 by 1447.861.8+14
- HLE (with tools)Kimi K2.6 by 10.843.254+10.8
- frontiermath_tier_4_v1Kimi K2.6 by 10.44.214.6+10.4
- livebench_data_analysisClaude Opus 4.5 by 9.374.465.1+9.3
- AA-OmniscienceClaude Opus 4.5 by 8.7145.3+8.7
- Terminal-Bench 2.0Kimi K2.6 by 7.459.366.7+7.4
- DeepSearchQA (F1)Kimi K2.6 by 6.976.183+6.9
- ToolathlonKimi K2.6 by 6.543.550+6.5
- τ²-Bench Telecom (AA run)Kimi K2.6 by 6.489.595.9+6.4
- livebench_languageClaude Opus 4.5 by 6.281.375.1+6.2
- WideSearch (item-f1)Kimi K2.6 by 4.676.280.8+4.6
- LiveBenchClaude Opus 4.5 by 3.87672.2+3.8
- LiveCodeBenchKimi K2.6 by 2.68789.6+2.6
- MMLU-ProClaude Opus 4.5 by 2.489.587.1+2.4
- AA IntelligenceClaude Opus 4.5 by 2.129.127+2.1
- vectara_answer_rateKimi K2.6 by 198.799.7+1
- vectara_avg_summary_lengthtie114.5116.7tie
- vectara_factual_consistencytie89.189.2tie
Questions people ask
Which is better, Claude Opus 4.5 or Kimi K2.6?
Kimi K2.6 wins seven of the eight areas we test: coding, agents, reasoning, images and charts, math, long documents and following instructions. Claude Opus 4.5 wins facts.
Which is better for coding?
Kimi K2.6. It wins 6 of the 9 coding tests both models report; Claude Opus 4.5 wins 2, and 1 is a tie.
Which is cheaper?
Claude Opus 4.5 costs $5.00 per million input tokens and $25.00 per million output tokens; Kimi K2.6 costs $0.95 and $4.00. That makes Kimi K2.6 about 84% cheaper for the same work.
How do you compare the two?
We use the 52 benchmark tests both models have published scores on. The verdict counts the 32 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 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.