Enter opens the highlighted result. Esc closes.

Claude Opus 5 vs Kimi K3

Updated October 21, 2026. 111 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Claude Opus 5 Replace Kimi K3
Claude Opus 5Anthropic
Coding
81.8
Overall
50.8
Output $/M
$25.00
97
Kimi K3Moonshot AI
Coding
61.8
Overall
43.6
Output $/M
$11.936
14
vs boards won of 111 shared

Decision reading

Pick Claude Opus 5 for coding: it scores 81.8 on the Vibe Coding Index against 61.8 for Kimi K3.

Pick Kimi K3 when output tokens cost 2.1x less, or when a solved task costs $0.81 against $1.33.

Model AClaude Opus 5Anthropic81.8/1002 independent benchmarks$1.33 per solved taskModel BKimi K3Moonshot AI61.8/1002 independent benchmarks$0.81 per solved task
Shared results
111
Claude Opus 5 only
19
Kimi K3 only
6
Like-for-like categories
8 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Claude Opus 5 81.8 from 2 benchmarks, Kimi K3 61.8 from 2 benchmarks.
Widest category
Agentic work, 12.4 points apart on average over 21 shared benchmarks.
Widest single gap
LiveBench browser: Claude Opus 5 84.0% against Kimi K3 16.0%, 68.0 points apart on a 0 to 100 scale.
Output price
Claude Opus 5 $25.00 against Kimi K3 $11.936 per million output tokens, 2.1x apart.
Context window
Claude Opus 5 takes 1M tokens, Kimi K3 1.05M.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryClaude Opus 5BarsKimi K3Ties
Coding212
Intelligence20
Agentic work183
Reasoning140
Math60
Knowledge101
Web design35
General233
All shared boards9714
CodingLike-for-likeFour of these boards feed the Vibe Coding Index212
Coding benchmarks for Claude Opus 5 and Kimi K3
BenchmarkClaude Opus 5Kimi K3Leader
LiveBench coding90.8%59.3%Claude Opus 5
LMArena WebDev16891655Claude Opus 5
AA Coding Index78.076.2Claude Opus 5
LiveCodeBench (Vals AI)89.0%87.2%Claude Opus 5
SWE-bench Verified (Vals AI)97.0%93.4%Claude Opus 5
Terminal-Bench 4 (Vals AI)53.5%17.2%Claude Opus 5
Vibe Code Bench (Vals AI)88.4%85.0%Claude Opus 5
Code migration (Vals AI)57.5%16.1%Claude Opus 5
ProgramBench (Vals AI)3.0%2.0%Claude Opus 5
IOI (Vals AI)84.3%48.9%Claude Opus 5
Terminal-Bench 2.1 (Vals AI)84.6%80.9%Claude Opus 5
SWE-bench Multimodal (Vals AI)Not measured44.2%Only Kimi K3 measured
Vibe Code Bench 1-100 (Vals AI)28.5%18.2%Claude Opus 5
SkillsBench (Vals AI)60.4%Not measuredOnly Claude Opus 5 measured
CyberBench (Vals AI)65.4%75.2%Kimi K3
SciCode56.4%59.5%Kimi K3
FrontierCode53.4%44.2%Claude Opus 5
FrontierSWE52.0%25.9%Claude Opus 5
WeirdML91.8%82.6%Claude Opus 5
ALE-Bench2164.61524.5Claude Opus 5
LMArena Text: coding15331508Claude Opus 5
LMArena Text: software and IT15271499Claude Opus 5
LMArena WebDev: React17111670Claude Opus 5
LMArena WebDev: HTML16801644Claude Opus 5
LMArena WebDev: from an image16621579Claude Opus 5
Neither model has a result on the other 13 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Claude Opus 5 and Kimi K3
BenchmarkClaude Opus 5Kimi K3Leader
AA Intelligence Index50.843.6Claude Opus 5
Epoch Capabilities Index163.9157.9Claude Opus 5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index183
Agentic work benchmarks for Claude Opus 5 and Kimi K3
BenchmarkClaude Opus 5Kimi K3Leader
AA Agentic Index56.550.0Claude Opus 5
Design Arena: agents building web apps12661296Kimi K3
Design Arena: agents building full-stack apps13061315Kimi K3
Design Arena: agents building mobile apps13481267Claude Opus 5
Design Arena: agents building Android apps12581256Claude Opus 5
Design Arena: agents building Godot games12791223Claude Opus 5
Design Arena: agents building games12611252Claude Opus 5
Design Arena: HTML slidesNot measured1256Only Kimi K3 measured
Design Arena: PowerPoint slides12651273Kimi K3
Design Arena: spreadsheetsNot measured1058Only Kimi K3 measured
LiveBench research88.9%79.0%Claude Opus 5
LiveBench browser84.0%16.0%Claude Opus 5
LiveBench long agentic62.1%42.8%Claude Opus 5
SREBench (Vals AI)12.2%Not measuredOnly Claude Opus 5 measured
Finance Agent (Vals AI)58.6%53.1%Claude Opus 5
Tax Agent Bench (Vals AI)45.5%33.4%Claude Opus 5
Legal Research Bench (Vals AI)55.3%46.1%Claude Opus 5
Harvey's Legal Agent Benchmark (Vals AI)Not measured12.9%Only Kimi K3 measured
Excel Modeling Benchmark (Vals AI)73.6%66.7%Claude Opus 5
BioMysteryBench (Vals AI)79.3%72.6%Claude Opus 5
Terminal-Bench-Science (Vals AI)27.1%1.4%Claude Opus 5
Vals Science Index47.9%Not measuredOnly Claude Opus 5 measured
Time Horizon Index (Vals AI)18.8%10.5%Claude Opus 5
CUA-bench (Vals AI)9.0%Not measuredOnly Claude Opus 5 measured
APEX-Agents65.8%47.8%Claude Opus 5
OSWorld 2.031.4%Not measuredOnly Claude Opus 5 measured
BALROG63.4%Not measuredOnly Claude Opus 5 measured
PostTrainBench35.0%32.0%Claude Opus 5
Vending-Bench 2$11,182$5,165Claude Opus 5
LMArena Search1242Not measuredOnly Claude Opus 5 measured
Neither model has a result on the other 9 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index140
Reasoning benchmarks for Claude Opus 5 and Kimi K3
BenchmarkClaude Opus 5Kimi K3Leader
GPQA Diamond93.4%92.9%Claude Opus 5
LiveBench reasoning91.2%90.7%Claude Opus 5
MysteryMechanism (Vals AI)37.4%Not measuredOnly Claude Opus 5 measured
MMMU Pro (Vals AI)89.9%88.2%Claude Opus 5
GPQA Diamond (Artificial Analysis run)93.7%93.5%Claude Opus 5
Humanity's Last Exam (Artificial Analysis run)54.9%46.9%Claude Opus 5
GPQA Diamond (Epoch AI run)93.9%93.1%Claude Opus 5
ARC-AGI97.5%94.5%Claude Opus 5
ARC-AGI-290.4%60.4%Claude Opus 5
SimpleBench80.6%60.7%Claude Opus 5
Chess puzzles43.0%37.0%Claude Opus 5
Mystery game puzzles59.0%26.0%Claude Opus 5
EBR-bench45.7%Not measuredOnly Claude Opus 5 measured
LMArena Text: hard prompts15271495Claude Opus 5
LMArena Text: expert15541520Claude Opus 5
LMArena Text: hard prompts in English15231496Claude Opus 5
LMArena Vision: diagrams1348Not measuredOnly Claude Opus 5 measured
LMArena Vision: homework1338Not measuredOnly Claude Opus 5 measured
Neither model has a result on the other 3 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index60
Math benchmarks for Claude Opus 5 and Kimi K3
BenchmarkClaude Opus 5Kimi K3Leader
OTIS Mock AIME98.9%97.2%Claude Opus 5
FrontierMath, tiers 1 to 385.6%72.2%Claude Opus 5
FrontierMath, tier 473.2%39.0%Claude Opus 5
ProofBench99.0%87.0%Claude Opus 5
LMArena Text: math15331494Claude Opus 5
LMArena Text: mathematical work15441505Claude Opus 5
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index101
Knowledge benchmarks for Claude Opus 5 and Kimi K3
BenchmarkClaude Opus 5Kimi K3Leader
MMLU Pro91.6%88.0%Claude Opus 5
LegalBench (Vals AI)87.0%86.2%Claude Opus 5
CorpFin (Vals AI)73.2%71.6%Claude Opus 5
TaxEval (Vals AI)75.1%75.7%Kimi K3
MortgageTax (Vals AI)72.1%66.3%Claude Opus 5
MedCode (Vals AI)63.6%49.4%Claude Opus 5
SimpleQA Verified59.9%50.6%Claude Opus 5
LMArena Text: science15311508Claude Opus 5
LMArena Text: medicine and healthcare15251489Claude Opus 5
LMArena Text: legal and government15301503Claude Opus 5
LMArena Text: business and finance15011470Claude Opus 5
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index35
Web design benchmarks for Claude Opus 5 and Kimi K3
BenchmarkClaude Opus 5Kimi K3Leader
Design Arena: websites13091336Kimi K3
Design Arena: UI components13481345Claude Opus 5
Design Arena: game dev13391361Kimi K3
Design Arena: data viz13291350Kimi K3
Design Arena: 3D13351388Kimi K3
Design Arena: all code categories13241361Kimi K3
Design Arena: SVG13101301Claude Opus 5
Design Arena: ASCII art1352Not measuredOnly Claude Opus 5 measured
LiveBench design83.0%77.2%Claude Opus 5
GeneralLike-for-likePublic boards, shown as published, kept out of the index233
General benchmarks for Claude Opus 5 and Kimi K3
BenchmarkClaude Opus 5Kimi K3Leader
Vals IndexNot measured53.0%Only Kimi K3 measured
LiveBench data analysis74.6%78.7%Kimi K3
Public Benefits Bench (Vals AI)76.9%68.2%Claude Opus 5
SAGE (Vals AI)49.4%52.8%Kimi K3
MedScribe (Vals AI)91.0%88.0%Claude Opus 5
Vals Multimodal Index73.9%73.4%Claude Opus 5
Long context reasoning (AA-LCR)82.0%88.7%Kimi K3
DTBench97.9%91.2%Claude Opus 5
LMCA64.5%52.7%Claude Opus 5
Surface Evolver BenchNot measured95.0%Only Kimi K3 measured
Furniture assembly60.8%34.2%Claude Opus 5
LMArena Text15091476Claude Opus 5
LMArena Text: instruction following15181485Claude Opus 5
LMArena Text: creative writing14931457Claude Opus 5
LMArena Vision1320Not measuredOnly Claude Opus 5 measured
LMArena Text: English15081481Claude Opus 5
LMArena Text: non-English15021466Claude Opus 5
LMArena Text: Chinese15731529Claude Opus 5
LMArena Text: German15271490Claude Opus 5
LMArena Text: Japanese15201494Claude Opus 5
LMArena Text: Korean15171460Claude Opus 5
LMArena Text: French15191492Claude Opus 5
LMArena Text: Spanish15231472Claude Opus 5
LMArena Text: Russian15071485Claude Opus 5
LMArena Text: Polish15221458Claude Opus 5
LMArena Text: multi-turn15001489Claude Opus 5
LMArena Text: longer queries15171494Claude Opus 5
LMArena Text: writing and language14991471Claude Opus 5
LMArena Text: entertainment and media14811453Claude Opus 5
LMArena Vision: English1315Not measuredOnly Claude Opus 5 measured
LMArena Vision: Chinese1400Not measuredOnly Claude Opus 5 measured
LMArena Vision: reading text1329Not measuredOnly Claude Opus 5 measured
LMArena Vision: humor1293Not measuredOnly Claude Opus 5 measured
LMArena Vision: creative writing1313Not measuredOnly Claude Opus 5 measured
LMArena Document1516Not measuredOnly Claude Opus 5 measured
Neither model has a result on the other 11 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Opus 5 and Kimi K3
BenchmarkClaude Opus 5Kimi K3Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Claude Opus 5$0.018Fits in one request
Kimi K3$0.0062Fits in one request

Kimi K3 costs less

Repository review

50K input + 3K output tokens

Claude Opus 5$0.325Fits in one request
Kimi K3$0.048Fits in one request

Kimi K3 costs less

Agent loop

200K input + 10K output tokens

Claude Opus 5$1.25Fits in one request
Kimi K3$0.167Fits in one request

Kimi K3 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Claude Opus 5 and Kimi K3
SpecificationAnthropic Claude Opus 5MoonshotAI Kimi K3
MakerAnthropicMoonshot AI
WeightsNo public weights listedOpen weights
ReleasedJuly 24, 2026July 16, 2026
Context window1M1.05M
Input $/Mtok$5.00$0.239
Output $/Mtok$25.00$11.936
Coding index81.861.8
Independent measurements behind it22
Cost per solved task$1.33$0.81
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Opus 5 better than Kimi K3 for coding?

Pick Claude Opus 5 for coding: it scores 81.8 on the Vibe Coding Index against 61.8 for Kimi K3. Pick Kimi K3 when output tokens cost 2.1x less, or when a solved task costs $0.81 against $1.33.

Which is cheaper, Claude Opus 5 or Kimi K3?

Claude Opus 5 costs $5.00 per million input tokens and $25.00 per million output tokens; Kimi K3 costs $0.239 and $11.936. Per solved SWE-bench Verified task, Claude Opus 5 costs $1.33 and Kimi K3 $0.81.

Which has the bigger context window, Claude Opus 5 or Kimi K3?

Kimi K3, with 1.05M tokens against 1M.

How many benchmarks is the Claude Opus 5 vs Kimi K3 comparison based on?

111 benchmarks have a result for both models as of October 21, 2026. Claude Opus 5's coding index rests on 2 independent benchmarks and Kimi K3's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/claude-opus-5-vs-kimi-k3" title="Claude Opus 5 vs Kimi K3, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/claude-opus-5-vs-kimi-k3">LLMBenchmark.ai, Claude Opus 5 vs Kimi K3</a></p>