Kimi K2.6 vs Qwen3.6 Plus
Wins 7 of 8 areas
Coding · Agents · Reasoning · Images and charts · Math · Long documents · Following instructions
Wins 1 of 8 areas
Facts
Kimi K2.6 is the stronger all-rounder.Qwen3.6 Plus is cheaper and better at facts.
Scores updated · 57 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 software80Kimi K2.68 of 10 tests · 2 ties
- MathCompetition and research-level math40Kimi K2.64 of 5 tests · 1 tie
- ReasoningHard problems that need careful thinking40Kimi K2.64 of 4 tests
- AgentsCarrying out multi-step tasks on its own31Kimi K2.63 of 4 tests
- Images and chartsUnderstanding pictures, charts and video20Kimi K2.62 of 2 tests
- Following instructionsDoing exactly what it is asked10Kimi K2.61 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts10Kimi K2.61 of 1 test
- FactsGetting facts right instead of making them up12Qwen3.6 Plus2 of 3 tests
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.
Qwen3.6 Plus costs 29% 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
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+18.9points ahead
- Code for real scientific research problemsSciCode+10.8points ahead
- Completes tasks by operating a computer desktopOSWorld-Verified+10.6points ahead
Where Qwen3.6 Plus pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+9.2points ahead
- Multi-step tasks using many real software toolsMCP Atlas+6points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+5.9points ahead
Every test, side by side
All 57 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingKimi K2.6
- SciCodeKimi K2.6 by 10.851.540.7+10.8
- LiveBench · Agentic CodingKimi K2.6 by 5.546.941.4+5.5
- LMArena · WebDevKimi K2.6 by 48 rating points15091461+48 rating
- Terminal-Bench 2.1Kimi K2.6 by 4.565.961.4+4.5
- SWE-bench MultilingualKimi K2.6 by 2.976.773.8+2.9
- LiveCodeBench v6Kimi K2.6 by 2.589.687.1+2.5
- SWE-bench ProKimi K2.6 by 258.656.6+2
- SWE-bench VerifiedKimi K2.6 by 1.480.278.8+1.4
- LiveBench · Codingtie78.678.2tie
- Terminal-Bench Hardtie43.943.9tie
AgentsKimi K2.6
- OSWorld-VerifiedKimi K2.6 by 10.673.162.5+10.6
- MCP AtlasQwen3.6 Plus by 668.174.1+6
- τ-Bench V3 · BankingKimi K2.6 by 2.523.320.8+2.5
- GDPValKimi K2.6 by 2.32724.7+2.3
ReasoningKimi K2.6
- Humanity's Last ExamKimi K2.6 by 9.737.527.8+9.7
- CritPtKimi K2.6 by 5.182.9+5.1
- LiveBench · ReasoningKimi K2.6 by 3.679.475.8+3.6
- GPQA DiamondKimi K2.6 by 2.991.188.2+2.9
FactsQwen3.6 Plus
- SimpleQA VerifiedQwen3.6 Plus by 9.234.944.1+9.2
- AA-Omniscience · AccuracyKimi K2.6 by 6.232.626.4+6.2
- AA-Omniscience · Non-hallucinationQwen3.6 Plus by 5.959.565.4+5.9
Images and chartsKimi K2.6
- CharXiv (RQ)Kimi K2.6 by 5.286.781.5+5.2
- MMMU-ProKimi K2.6 by 1.479.478+1.4
MathKimi K2.6
- FrontierMath Tiers 1-3 (v2)Kimi K2.6 by 18.957.238.3+18.9
- HMMT Feb. 2026Kimi K2.6 by 4.992.787.8+4.9
- IMOAnswerBenchKimi K2.6 by 2.28683.8+2.2
- AIME 2026Kimi K2.6 by 1.196.495.3+1.1
- LiveBench · Mathematicstie84.383.7tie
Following instructionsKimi K2.6
- LiveBench · Instruction FollowingKimi K2.6 by 6.164.458.3+6.1
- IFBenchtie7675.2tie
Other results26 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.
- Claw Eval (pass@3)Kimi K2.6 by 22.280.958.7+22.2
- ToolathlonKimi K2.6 by 10.25039.8+10.2
- AgentWorldBench - MCPKimi K2.6 by 9.965.255.3+9.9
- MCPMarkKimi K2.6 by 7.755.948.2+7.7
- Artificial Analysis Coding IndexKimi K2.6 by 7.361.854.5+7.3
- AA Agentic IndexQwen3.6 Plus by 6.922.129+6.9
- WideSearchKimi K2.6 by 6.580.874.3+6.5
- frontiermath_tier_4_v1Kimi K2.6 by 6.314.68.3+6.3
- AgentWorldBench - SearchKimi K2.6 by 5.627.521.9+5.6
- MATH-VisionKimi K2.6 by 5.293.288+5.2
- Terminal-Bench 2.0Kimi K2.6 by 5.166.761.6+5.1
- livebench_data_analysisQwen3.6 Plus by 4.865.169.9+4.8
- AA-OmniscienceKimi K2.6 by 4.45.30.9+4.4
- Finance Agent v2Kimi K2.6 by 444.940.9+4
- HLE (with tools)Kimi K2.6 by 3.45450.6+3.4
- AgentWorldBench - OverallKimi K2.6 by 2.653.450.8+2.6
- AgentWorldBench - TerminalKimi K2.6 by 1.952.550.6+1.9
- τ²-Bench Telecom (AA run)Qwen3.6 Plus by 1.895.997.7+1.8
- MMLU-ProQwen3.6 Plus by 1.487.188.5+1.4
- LiveBenchKimi K2.6 by 1.372.270.8+1.3
- AgentWorldBench - AndroidKimi K2.6 by 1.258.957.6+1.2
- AgentWorldBench - Webtie50.250.8tie
- AgentWorldBench - OStie60.860.3tie
- AgentWorldBench - SWEtie58.859.1tie
- livebench_languagetie75.175tie
- AA Intelligencetie2727tie
Questions people ask
Which is better, Kimi K2.6 or Qwen3.6 Plus?
Kimi K2.6 wins seven of the eight areas we test: coding, agents, reasoning, images and charts, math, long documents and following instructions. Qwen3.6 Plus wins facts, and costs 29% less.
Which is better for coding?
Kimi K2.6. It wins 8 of the 10 coding tests both models report; Qwen3.6 Plus wins none, and 2 are ties.
Which is cheaper?
Kimi K2.6 costs $0.95 per million input tokens and $4.00 per million output tokens; Qwen3.6 Plus costs $0.50 and $3.00. That makes Qwen3.6 Plus about 29% cheaper for the same work.
How do you compare the two?
We use the 57 benchmark tests both models have published scores on. The verdict counts the 31 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 26 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.