BenchLeader

Kimi K3.1 Pro vs o3

Full profiles: Kimi K3.1 Pro (Moonshot AI) and o3 (OpenAI): every benchmark, price and rank.

Updated · Data from 15 sources

As of 16 Nov 2026, Kimi K3.1 Pro and o3 are neck and neck on quality (61.8 vs 60.4 on the BenchLeader Index, inside the ±4.4 margin), and they cost about the same at $4.00 vs $3.50 per 1M tokens blended. Choose Kimi K3.1 Pro for multimodal (66 vs 53), long context (68 vs 62), and maths (73 vs 71); choose o3 for fast first answers (5.2 s vs 157.4 s) and agents & tools (70 vs 65) and reasoning (61 vs 59), where it also streams 3×+ faster.

Key facts (as of 16 Nov 2026)

  • Kimi K3.1 Pro scores 61.8 on the BenchLeader Index and o3 scores 60.4 (BenchLeader, 16 Nov 2026).
  • Kimi K3.1 Pro costs $4.00 per 1M tokens blended (list price, 16 Nov 2026).
  • o3 costs $3.50 per 1M tokens blended (list price, 16 Nov 2026).
  • o3 streams 3×+ faster than Kimi K3.1 Pro (Artificial Analysis, 7-day median, 16 Nov 2026).
  • Of the 13 benchmarks both models have, Kimi K3.1 Pro leads 10 (BenchLeader, 16 Nov 2026).

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
CodingEither60 vs 60 category score; GameBench 2 Overall 78.3 vs 34.1Either
Agents & tool useo370 vs 65 category scoreMedium
MathsKimi K3.1 Pro73 vs 71 category scoreMedium
Reasoningo361 vs 59 category score; Humanity's Last Exam (AA) 20.1% vs 52.5%Medium
KnowledgeKimi K3.1 Pro58 vs 56 category score; PRBench Finance 47.5% vs 47.7%Medium
Long documentsKimi K3.1 Pro68 vs 62 category score; AA-LCR v1.1 86.3% vs 74.7%Medium
Images and other mediaKimi K3.1 Pro66 vs 53 category score; MMMU-Pro 83.9% vs 70.1%High
Chat quality (human preference)–Only o3 has an Arena rating (61 category score)Not comparable
Lowest costEither$4.00 vs $3.50 per 1M blended; agentic coding session $0.160 vs $0.152–
Fastest first answero35.2 s (o3) vs 157.4 s (Kimi K3.1 Pro) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • Kimi K3.1 Pro and o3 are neck and neck on quality (61.8 vs 60.4, within the ±4.4 margin).
  • Kimi K3.1 Pro is stronger in knowledge, long context, maths, multimodal.
  • o3 is stronger in agents & tools, coding, reasoning.
  • They cost about the same ($3.50 per 1M blended).
  • o3 streams 4.4× faster (131 vs 30 tokens per second).

At a glance

SpecKimi K3.1 Proo3
DeveloperMoonshot AIOpenAI
Released23 Oct 202616 Apr 2025
Context window (tokens)1,048,576200,000
Compared atKimi K3.1 Proo3
Open weightsYesNo
Price in / out / cached input per 1M$2.00 / $10.00 / $0.200$2.00 / $8.00 / $0.500

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

CategoryKimi K3.1 Proo3Leader
Reasoning59.260.9o3
Coding59.659.8level
Agents & tools64.670.3o3
Maths73.470.6Kimi K3.1 Pro
Knowledge58.256.2Kimi K3.1 Pro
Instruction following–65.2not comparable
Human preference–61.0not comparable
Multimodal66.352.8Kimi K3.1 Pro
Long context67.961.9Kimi K3.1 Pro

On one chart

Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.

2550 avg75ReasoningCodingAgents & toolsMathsKnowledgeLong contextMultimodalKimi K3.1 Pro, Reasoning: 59.2Kimi K3.1 Pro, Coding: 59.6Kimi K3.1 Pro, Agents & tools: 64.6Kimi K3.1 Pro, Maths: 73.4Kimi K3.1 Pro, Knowledge: 58.2Kimi K3.1 Pro, Long context: 67.9Kimi K3.1 Pro, Multimodal: 66.3o3, Reasoning: 60.9o3, Coding: 59.8o3, Agents & tools: 70.3o3, Maths: 70.6o3, Knowledge: 56.2o3, Long context: 61.9o3, Multimodal: 52.8

Not drawn, measured for fewer than two of these models: Instruction following, Human preference.

Model, click to add or removeIndex

Where sources disagree

Across the 13 benchmarks both have, Kimi K3.1 Pro leads 10, o3 leads 3. Every overall index that rates both puts Kimi K3.1 Pro ahead.

Overall indexKimi K3.1 Proo3Leader
BenchLeader Index61.860.4neck and neck
Arena Text1485†1432Kimi K3.1 Pro

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.

MeasureKimi K3.1 Proo3
Blended price per 1M tokens$4.00$3.50
Input / output per 1M$2.00 / $10.00$2.00 / $8.00
Cached input per 1M$0.200$0.500
Price per task: chat reply$0.0038$0.0032
Price per task: summarize a 30-page report$0.030$0.029
Price per task: code edit$0.027$0.024
Price per task: agentic coding session$0.160$0.152
Price per task: structured extraction$0.0060$0.0056
Agentic coding session, input mostly cached$0.079$0.085
Cost per benchmark task (AA Index)$3.68–
Output tokens per task (AA Index)72k–
Of which reasoning44k–
Time per task (AA Index)40 min–
Output speed30 tok/s131 tok/s
Time to first answer157.4 s5.2 s
Time to a complete answer, last measured 13 Nov 2026417.5 s17.3 s

Each model is compared at its best-ranked reasoning-effort setting. Kimi K3.1 Pro scores 61.8 and takes 157 s to its first answer token; Kimi K3.1 Pro (max) scores 59.9 in 157 s. o3 scores 60.4 and takes 5 s to its first answer token; o3 (high) scores 55.1 in 5 s, o3 (medium) scores 53.3 in 5 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

161 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublisherKimi K3.1 Proo3
BenchLeader IndexBenchLeader61.860.4
Agents & tools scoreBenchLeader64.670.3
Coding scoreBenchLeader59.659.8
Knowledge scoreBenchLeader58.256.2
Long context scoreBenchLeader67.961.9
Maths scoreBenchLeader73.470.6
Multimodal scoreBenchLeader66.352.8
Reasoning scoreBenchLeader59.260.9
Human preference scoreBenchLeader–61.0
Instruction following scoreBenchLeader–65.2
Blended price $/MBenchLeader$4.00$3.50
Output speedBenchLeader30 tok/s131 tok/s
Time to first answerBenchLeader157.4 s5.2 s
Context windowBenchLeader1,048,576 tokens200,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub–81.8%†
Humanity's Last ExamScale AI / CAIS–20.3%†
SimpleBenchSimpleBench–53.1%†
Arena Hard PromptsArena (LMArena)1504†1441
LiveBench 2.0 ReasoningLiveBench91.1%†–
GPQA Diamond (AA)Artificial Analysis–82.7%
Humanity's Last Exam (AA)Artificial Analysis52.5%20.1%
HLE-Diamond (AA)Artificial Analysis37.3%–
GPQA Diamond (Vals)Vals AI–84.1%†
Kagi LLM BenchmarkKagi LLM Benchmark–67.6%
GPQA Diamond (HELM)HELM Capabilities–75.3%
ARC-AGI-1ARC Prize–60.8%†
ARC-AGI-2ARC Prize–6.5%†
CritPtArtificial Analysis24.6%1.1%
CritPt v2 hard subset (AA)Artificial Analysis49.5%–
MysteryMechanismVals AI26.1%†–
SciPredictScale AI SEAL–17.9%
Arena ExpertArena (LMArena)1515†1448
Chess PuzzlesEpoch AI Benchmarking Hub–36.0%†
Mystery Game PuzzlesEpoch AI Benchmarking Hub–29.0%†
CL-benchEpoch AI Benchmarking Hub–17.8%†
LMCAEpoch AI Benchmarking Hub57.6%†39.7%†
DTBenchEpoch AI Benchmarking Hub93.3%†84.8%†
ForecastBenchForecastBench–64.3%
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub–62.3%†
Show all 161 benchmarks
MetricPublisherKimi K3.1 Proo3
Coding
SciCodeSciCode60.8%–
WeirdMLWeirdML–52.4%†
GSO-BenchGSO-Bench–8.8%†
Arena CodingArena (LMArena)1530†1460
Arena WebDevArena (LMArena)1687†–
LiveBench 2.0 CodingLiveBench67.8%†–
LiveCodeBenchVals AI–83.9%†
IOIVals AI81.9%†–
Aider PolyglotAider polyglot leaderboard–76.9%
SWE-bench Verified (bash only)SWE-bench–58.4%
SWE-bench Verified (any scaffold)SWE-bench–58.4%
Vibe Code Bench v1.1Vals AI84.9%†–
SWE Atlas v1.1Scale AI SEAL61.1%–
SWE Atlas v1.1: Codebase QnAScale AI SEAL70.2%–
SWE Atlas v1.1: Test WritingScale AI SEAL58.9%–
SWE Atlas v1.1: RefactoringScale AI SEAL54.3%–
CadEvalEpoch AI Benchmarking Hub–74.0%†
ALE-BenchEpoch AI Benchmarking Hub1932.8†933.5†
WeirdML v3WeirdML9.9%†–
Terminal-Bench 4.0 (AA)Artificial Analysis67.2%–
GameBench 2 OverallDuelLab GameBench 278.334.1
GameBench 2 by reasoning settingDuelLab GameBench 284.0†30.4†
Design Arena WebsiteDesign Arena13211028
FrontierCode 1.1 (AA)Artificial Analysis47.1%–
FrontierCode 2.0 (AA)Artificial Analysis25.4%–
LiveCodeBench (AA)Artificial Analysis–80.8%
Agents & tools
Terminal-BenchTerminal-Bench59.7%†–
GDPvalOpenAI–30.8%†
APEX-AgentsMercor65.5%–
Arena AgentArena (LMArena)4.9†–
LiveBench Long AgenticLiveBench38.7%†–
LiveBench BrowserLiveBench75.9%†–
Terminal-Bench HardArtificial Analysis–37.1%
τ²-Bench Telecom (AA)Artificial Analysis–80.7%
HiL-BenchScale AI SEAL61.3%–
BFCL OverallBerkeley Function Calling Leaderboard–63.0%
GDPval-AA v2.1, normalised scoreArtificial Analysis67.3%–
GDPval-AA v2.1Artificial Analysis1861–
Analyst Agent (AA)Artificial Analysis38.8%–
Finance Agent v2.1Vals AI61.5%†–
Harvey's Legal Agent Benchmark v1.1Vals AI11.7%†–
Legal Research BenchVals AI37.5%†–
SREBenchVals AI27.1%†–
Tax Agent BenchVals AI24.9%†–
Terminal-Bench 4.0 (Vals)Vals AI63.6%†–
Terminal-Bench ScienceVals AI31.4%†–
Arena SearchArena (LMArena)–1197
Arena Agent Task CompletionArena (LMArena)2.7†–
DeepResearch BenchDeepResearch Bench–45.2%†
METR Time HorizonsMETR–63.6%
AutomationBenchArtificial Analysis73.3%–
GDP.pdf (AA)Artificial Analysis31.4%–
AA-Briefcase v1.1Artificial Analysis1674–
Design Arena AgenticDesign Arena1301–
Terminal-Bench-Science 0.1 (AA)Artificial Analysis34.3%–
CWE-Bench (AA)Artificial Analysis60.8%–
DeepSec-Bench (AA)Artificial Analysis36.7%–
CyberGym end-to-end (AA)Artificial Analysis94.7%–
MKA (AA)Artificial Analysis43.9%–
BrowseComp-PlusBrowseComp-Plus–50.5%
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub–33.3%†
OTIS Mock AIMEEpoch AI Benchmarking Hub–83.9%†
MATH Level 5Epoch AI Benchmarking Hub–97.8%†
ProofBenchVals AI91.0%–
AIME (Vals)Vals AI–85.3%†
MGSMVals AI–91.8%†
IMO 2025MathArena–16.7%†
ArXivMathMathArena58.3%†–
BrokenArXivMathArena61.9%†–
Omni-MATH (HELM)HELM Capabilities–71.4%
MATH 500Vals AI–94.6%†
Arena MathsArena (LMArena)–1448
AIME 2025 (AA)Artificial Analysis–88.3%
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub–49.4%†
LiveBench 2.0 Data AnalysisLiveBench79.1%†–
LiveBench ResearchLiveBench82.1%†–
AA-Omniscience v1.1Artificial Analysis29.9–
MMLU-ProVals AI–85.6%†
LegalBenchVals AI–83.8%†
CorpFinVals AI–59.7%†
TaxEvalVals AI–74.6%†
MedQAVals AI–96.1%†
PRBench FinanceScale AI SEAL47.5%47.7%
PRBench LegalScale AI SEAL47.9%48.6%
MultiNRCScale AI SEAL–45.5%†
MMLU-Pro (HELM)HELM Capabilities–85.9%
BioMysteryBenchVals AI75.6%†–
Excel Modeling Benchmark v1.1Vals AI70.0%†–
MedCodeVals AI–47.3%†
MedScribeVals AI–76.7%†
Access to Justice BenchmarkVals AI69.5%†–
DrugDiscoveryBenchScale AI SEAL63.4%†–
AA-Omniscience v1.1: accuracyArtificial Analysis48.9%–
AA-Omniscience v1.1: non-hallucinationArtificial Analysis62.7%–
Instruction following
IFBenchArtificial Analysis–71.4%
MultiChallengeScale AI SEAL–56.6%†
TutorBenchScale AI SEAL–52.1%†
IFEval (HELM)HELM Capabilities–86.9%
WildBench (HELM)HELM Capabilities–86.1%
Arena Instruction FollowingArena (LMArena)1487†1403
Human preference
Arena TextArena (LMArena)1485†1432
EQ-Bench Creative Writing v3EQ-Bench21731675
Arena Creative WritingArena (LMArena)1453†1383
Arena Multi-turnArena (LMArena)1475†1420
Arena Longer QueriesArena (LMArena)1491†1411
Lech Mazur Creative WritingLech Mazur–83.9%†
Multimodal
Arena VisionArena (LMArena)–1214
LiveBench DesignLiveBench73.7%†–
MMMU-ProArtificial Analysis83.9%70.1%
VISTAScale AI SEAL–50.1%†
MMMU-Pro (Vals)Vals AI–80.4%†
MortgageTaxVals AI–65.7%†
Humanity's Sixth SenseScale AI SEAL33.2%†–
VTBScale AI SEAL–13.7%
GeoBenchGeoBench–60.0%†
VPCTEpoch AI Benchmarking Hub–52.0%†
VIALS-AAHandshake AI45.3%–
AA-3DArtificial Analysis1112–
Long context
Fiction.LiveBench 120kFiction.live–100.0%†
AA-LCR v1.1Artificial Analysis86.3%74.7%
MLCRArtificial Analysis38.3%–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub161.4147.1
LiveBench 2.0LiveBench68.7%†–
AA Intelligence Index v5.0.1Artificial Analysis46.5–
AA Intelligence Index v4.3.2Artificial Analysis55.120.2
AA Progress Index v1Artificial Analysis340.7258.3
Vals IndexVals AI62.5†–
HELM Capabilities meanHELM Capabilities–81.1%
Vals Finance IndexVals AI56.0%†–
Vals Science IndexVals AI44.4%†–
Vals Legal IndexVals AI67.0%†–
AA Cyber IndexArtificial Analysis64.1–
Safety & honesty
FORTRESSScale AI SEAL18.6%16.0%
MASKScale AI SEAL–84.5%†
PropensityBenchScale AI SEAL–10.5%
Metadata
AA Openness IndexArtificial Analysis33.35.6

Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • Kimi K3.1 Pro: BenchLeader Index 61.2 → 61.8.
  • o3: BenchLeader Index 61.2 → 60.4.
  • Kimi K3.1 Pro: 1 score revision.
  • o3: 1 score revision.
  • Kimi K3.1 Pro: 1 score revision.
  • Kimi K3.1 Pro: 1 new result (HiL-Bench).
  • Kimi K3.1 Pro: 3 new results (PRBench Finance, PRBench Legal, FORTRESS); 2 score revisions.
  • Kimi K3.1 Pro: 2 new results (APEX-Agents, Epoch Capabilities Index).
  • Kimi K3.1 Pro: BenchLeader Index 61.8 → 61.2.
  • Kimi K3.1 Pro: 5 score revisions.

Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs

Kimi K3.1 Pro vs o3: questions

Is Kimi K3.1 Pro better than o3?
Kimi K3.1 Pro and o3 are neck and neck on quality (61.8 vs 60.4, within the ±4.4 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 16 Nov 2026, so check the category scores for your use.
Kimi K3.1 Pro vs o3: which is cheaper, and will that change?
o3 is cheaper today: $3.50 against $4.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Kimi K3.1 Pro or o3 better for coding?
They are level in coding (60 vs 60 on the category index).
Which is faster, Kimi K3.1 Pro or o3?
o3 streams 3×+ faster (131 vs 30 output tokens per second, Artificial Analysis 7-day median).
Is Kimi K3.1 Pro or o3 better for agents and tool use?
o3 scores higher in agentic tasks (70 vs 65 on the category index, where 50 is average).
More questions (4)
Which is better for long documents?
Kimi K3.1 Pro scores higher on long-context benchmarks (68 vs 62). Kimi K3.1 Pro accepts more context: 1,048,576 against 200,000 tokens for o3.
Which has lower latency?
o3 answers first: 5.2 s against 157.4 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Do independent benchmarks agree on Kimi K3.1 Pro vs o3?
Across the 13 benchmarks both have, Kimi K3.1 Pro leads 10, o3 leads 3. Every overall index that rates both puts Kimi K3.1 Pro ahead.
When was Kimi K3.1 Pro released?
Kimi K3.1 Pro was released on 23 Oct 2026.