Claude Opus 4.6 vs Kimi K2 Instruct
Wins 6 of 7 areas
Coding · Agents · Reasoning · Facts · Math · Long documents
Wins 0 of 7 areas
—
Claude Opus 4.6 is the stronger all-rounder.Kimi K2 Instruct is cheaper.
Scores updated · 33 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.
- CodingWriting and fixing software60Claude Opus 4.66 of 6 tests
- ReasoningHard problems that need careful thinking30Claude Opus 4.63 of 3 tests
- FactsGetting facts right instead of making them up30Claude Opus 4.63 of 3 tests
- AgentsCarrying out multi-step tasks on its own20Claude Opus 4.62 of 2 tests
- MathCompetition and research-level math10Claude Opus 4.61 of 1 test
- Long documentsFinding answers in very long texts10Claude Opus 4.61 of 1 test
- Following instructionsDoing exactly what it is asked11Even1 each
Math and long documents 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 K2 Instruct costs 90% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Claude Opus 4.6 pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+69.9points ahead
- Real work tasks from 44 professionsGDPVal+37.7points ahead
- Recent programming contest problemsLiveCodeBench v6+35.1points ahead
Where Kimi K2 Instruct pulls ahead
- Keeps track of context across a multi-turn chatMulti-Challenge+16.9points ahead
Every test, side by side
All 33 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 4.6
- LiveCodeBench v6Claude Opus 4.6 by 35.188.853.7+35.1
- SWE-bench MultilingualClaude Opus 4.6 by 30.577.847.3+30.5
- SWE-bench ProClaude Opus 4.6 by 25.753.427.7+25.7
- Terminal-Bench HardClaude Opus 4.6 by 22.746.223.5+22.7
- SciCodeClaude Opus 4.6 by 21.251.930.7+21.2
- SWE-bench VerifiedClaude Opus 4.6 by 1580.865.8+15
AgentsClaude Opus 4.6
- BrowseCompClaude Opus 4.6 by 69.98414.1+69.9
- GDPValClaude Opus 4.6 by 37.755.918.2+37.7
ReasoningClaude Opus 4.6
- Humanity's Last ExamClaude Opus 4.6 by 33.539.96.4+33.5
- GPQA DiamondClaude Opus 4.6 by 12.989.676.7+12.9
- CritPtClaude Opus 4.6 by 12.612.60+12.6
FactsClaude Opus 4.6
- AA-Omniscience · AccuracyClaude Opus 4.6 by 21.64725.4+21.6
- AA-Omniscience · Non-hallucinationClaude Opus 4.6 by 6.937.230.3+6.9
- Vectara HHEM hallucination ratelower is betterClaude Opus 4.6 by 5.712.217.9+5.7
Following instructionsEven
- Multi-ChallengeKimi K2 Instruct by 16.937.154.1+16.9
- IFBenchClaude Opus 4.6 by 11.453.141.7+11.4
Other results15 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.
- AIME 2025Claude Opus 4.6 by 50.399.849.5+50.3
- AIME25 no toolsClaude Opus 4.6 by 44.695.651+44.6
- AA-OmniscienceClaude Opus 4.6 by 40.313.7-26.6+40.3
- LiveCodeBenchClaude Opus 4.6 by 35.188.853.7+35.1
- AA Agentic IndexClaude Opus 4.6 by 29.967.637.7+29.9
- HLE (with tools)Claude Opus 4.6 by 26.453.326.9+26.4
- Artificial Analysis Coding IndexClaude Opus 4.6 by 22.248.125.9+22.2
- τ²-Bench (Retail)Claude Opus 4.6 by 21.391.970.6+21.3
- τ²-Bench Telecom (AA run)Claude Opus 4.6 by 18.792.173.4+18.7
- AA IntelligenceClaude Opus 4.6 by 16.631.915.3+16.6
- MMLU-ProClaude Opus 4.6 by 889.181.1+8
- vectara_avg_summary_lengthClaude Opus 4.6 by 78.4 rating points137.659.2+78.4 rating
- vectara_factual_consistencyClaude Opus 4.6 by 5.787.882.1+5.7
- vectara_answer_rateClaude Opus 4.6 by 1.299.898.6+1.2
- LiveBenchtie76.376.4tie
Questions people ask
Which is better, Claude Opus 4.6 or Kimi K2 Instruct?
Claude Opus 4.6 wins six of the seven areas where both have results: coding, agents, reasoning, facts, math and long documents. Kimi K2 Instruct wins none, but costs 90% less. They are level on following instructions.
Which is better for coding?
Claude Opus 4.6. It wins 6 of the 6 coding tests both models report; Kimi K2 Instruct wins none.
Which is cheaper?
Claude Opus 4.6 costs $5.00 per million input tokens and $25.00 per million output tokens; Kimi K2 Instruct costs $0.60 and $2.50. That makes Kimi K2 Instruct about 90% cheaper for the same work.
How do you compare the two?
We use the 33 benchmark tests both models have published scores on. The verdict counts the 18 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 15 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.