Enter opens the highlighted result. Esc closes.

Kimi K2.5 vs Mistral Small 3

Kimi K2.5 leads Mistral Small 3 on the Artificial Analysis Intelligence Index, 23.5 to 6.7, and Mistral Small 3 costs 35.6x less per output token ($0.08 vs $2.85 per million). Updated October 21, 2026; 32 benchmarks have a result for both.

Replace Kimi K2.5 Replace Mistral Small 3
Kimi K2.5Moonshot AI
Coding
38.6
Overall
44.8
Output $/M
$2.85
32
Mistral Small 3Mistral AI
Coding
No data
Overall
No data
Output $/M
$0.08
0
vs boards won of 32 shared

Kimi K2.5 or Mistral Small 3: which to pick

Only Kimi K2.5 has a coding score, 38.6 on the Vibe Coding Index; Mistral Small 3 has none, so the two cannot be ranked on coding.

Pick Mistral Small 3 for output tokens 35.6x cheaper ($0.08 vs $2.85 per million).

Model AKimi K2.5Moonshot AI38.6/1003 independent benchmarks$0.24 per solved taskModel BMistral Small 3Mistral AINot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
32
Kimi K2.5 only
67
Mistral Small 3 only
4
Like-for-like categories
7 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

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
Kimi K2.5 38.6 from 3 benchmarks, Mistral Small 3 not benchmarked.
Widest category
Agentic work, 76.3 points apart on average over 1 shared benchmark.
Widest single gap
Long context reasoning (AA-LCR): Kimi K2.5 78.0% against Mistral Small 3 0.0%, 78.0 points apart on a 0 to 100 scale.
Output price
Kimi K2.5 $2.85 against Mistral Small 3 $0.08 per million output tokens, 35.6x apart.
Context window
Kimi K2.5 takes 262K tokens, Mistral Small 3 33K.

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
CategoryKimi K2.5BarsMistral Small 3Ties
Coding20
Intelligence10
Agentic work10
Reasoning50
Math20
Knowledge40
General170
All shared boards320
CodingLike-for-likeSix of these boards feed the Vibe Coding Index20
Coding benchmarks for Kimi K2.5 and Mistral Small 3
BenchmarkKimi K2.5Mistral Small 3Leader
SWE-bench Verified70.8%Not measuredOnly Kimi K2.5 measured
SWE-bench Multilingual67.3%Not measuredOnly Kimi K2.5 measured
LMArena WebDev1437Not measuredOnly Kimi K2.5 measured
AA Coding Index46.8Not measuredOnly Kimi K2.5 measured
LiveCodeBench (Vals AI)83.9%Not measuredOnly Kimi K2.5 measured
SWE-bench Verified (Vals AI)70.0%Not measuredOnly Kimi K2.5 measured
Vibe Code Bench (Vals AI)17.5%Not measuredOnly Kimi K2.5 measured
Code migration (Vals AI)7.0%Not measuredOnly Kimi K2.5 measured
Terminal-Bench 2.1 (Vals AI)42.0%Not measuredOnly Kimi K2.5 measured
LiveCodeBench (Artificial Analysis run)Not measured25.2%Only Mistral Small 3 measured
Terminal-Bench Hard (Artificial Analysis run)34.8%Not measuredOnly Kimi K2.5 measured
Terminal-Bench43.2%Not measuredOnly Kimi K2.5 measured
WeirdML45.6%Not measuredOnly Kimi K2.5 measured
ALE-Bench821.6Not measuredOnly Kimi K2.5 measured
LMArena Text: coding14831246Kimi K2.5
LMArena Text: software and IT14711248Kimi K2.5
LMArena WebDev: React1429Not measuredOnly Kimi K2.5 measured
LMArena WebDev: HTML1426Not measuredOnly Kimi K2.5 measured
LMArena WebDev: from an image1439Not measuredOnly Kimi K2.5 measured
Neither model has a result on the other 20 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for Kimi K2.5 and Mistral Small 3
BenchmarkKimi K2.5Mistral Small 3Leader
AA Intelligence Index23.56.7Kimi K2.5
Epoch Capabilities Index148.1Not measuredOnly Kimi K2.5 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for Kimi K2.5 and Mistral Small 3
BenchmarkKimi K2.5Mistral Small 3Leader
Design Arena: agents building web apps1136Not measuredOnly Kimi K2.5 measured
Design Arena: agents building full-stack apps1114Not measuredOnly Kimi K2.5 measured
Design Arena: agents building mobile apps1147Not measuredOnly Kimi K2.5 measured
Design Arena: agents building Android apps1118Not measuredOnly Kimi K2.5 measured
Design Arena: agents building Godot games1219Not measuredOnly Kimi K2.5 measured
Finance Agent (Vals AI)35.8%Not measuredOnly Kimi K2.5 measured
Legal Research Bench (Vals AI)15.9%Not measuredOnly Kimi K2.5 measured
Excel Modeling Benchmark (Vals AI)28.5%Not measuredOnly Kimi K2.5 measured
τ²-Bench Telecom95.9%19.6%Kimi K2.5
OSWorld63.3%Not measuredOnly Kimi K2.5 measured
Vending-Bench 2$1,198Not measuredOnly Kimi K2.5 measured
Neither model has a result on the other 30 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index50
Reasoning benchmarks for Kimi K2.5 and Mistral Small 3
BenchmarkKimi K2.5Mistral Small 3Leader
GPQA Diamond84.1%Not measuredOnly Kimi K2.5 measured
MMMU Pro (Vals AI)84.3%Not measuredOnly Kimi K2.5 measured
GPQA Diamond (Artificial Analysis run)87.9%46.2%Kimi K2.5
Humanity's Last Exam (Artificial Analysis run)30.7%3.8%Kimi K2.5
ARC-AGI65.3%Not measuredOnly Kimi K2.5 measured
ARC-AGI-211.8%Not measuredOnly Kimi K2.5 measured
SimpleBench46.8%Not measuredOnly Kimi K2.5 measured
LMArena Text: hard prompts14531233Kimi K2.5
LMArena Text: expert14661202Kimi K2.5
LMArena Text: hard prompts in English14601243Kimi K2.5
LMArena Vision: diagrams1282Not measuredOnly Kimi K2.5 measured
LMArena Vision: homework1297Not measuredOnly Kimi K2.5 measured
Kagi LLM Benchmark78.5%Not measuredOnly Kimi K2.5 measured
Neither model has a result on the other 9 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index20
Math benchmarks for Kimi K2.5 and Mistral Small 3
BenchmarkKimi K2.5Mistral Small 3Leader
AIME95.6%Not measuredOnly Kimi K2.5 measured
AA Math IndexNot measured4.3Only Mistral Small 3 measured
AIME 2025 (Artificial Analysis run)Not measured4.3%Only Mistral Small 3 measured
LMArena Text: math14701240Kimi K2.5
LMArena Text: mathematical work14731243Kimi K2.5
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
Knowledge benchmarks for Kimi K2.5 and Mistral Small 3
BenchmarkKimi K2.5Mistral Small 3Leader
MMLU Pro85.9%Not measuredOnly Kimi K2.5 measured
CaseLaw (Vals AI)58.7%Not measuredOnly Kimi K2.5 measured
CorpFin (Vals AI)68.3%Not measuredOnly Kimi K2.5 measured
TaxEval (Vals AI)74.2%Not measuredOnly Kimi K2.5 measured
MortgageTax (Vals AI)66.5%Not measuredOnly Kimi K2.5 measured
MedQA (Vals AI)94.4%Not measuredOnly Kimi K2.5 measured
MedCode (Vals AI)39.3%Not measuredOnly Kimi K2.5 measured
MMLU Pro (Artificial Analysis run)Not measured65.2%Only Mistral Small 3 measured
SimpleQA Verified34.3%Not measuredOnly Kimi K2.5 measured
LMArena Text: science14591235Kimi K2.5
LMArena Text: medicine and healthcare14481215Kimi K2.5
LMArena Text: legal and government14411219Kimi K2.5
LMArena Text: business and finance14271205Kimi K2.5
Neither model has a result on the other 5 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
Web design benchmarks for Kimi K2.5 and Mistral Small 3
BenchmarkKimi K2.5Mistral Small 3Leader
Design Arena: websites1248Not measuredOnly Kimi K2.5 measured
Design Arena: UI components1236Not measuredOnly Kimi K2.5 measured
Design Arena: game dev1207Not measuredOnly Kimi K2.5 measured
Design Arena: data viz1219Not measuredOnly Kimi K2.5 measured
Design Arena: 3D1205Not measuredOnly Kimi K2.5 measured
Design Arena: all code categories1236Not measuredOnly Kimi K2.5 measured
Design Arena: SVG1149Not measuredOnly Kimi K2.5 measured
Design Arena: ASCII art1166Not measuredOnly Kimi K2.5 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index170
General benchmarks for Kimi K2.5 and Mistral Small 3
BenchmarkKimi K2.5Mistral Small 3Leader
SAGE (Vals AI)49.9%Not measuredOnly Kimi K2.5 measured
MedScribe (Vals AI)76.4%Not measuredOnly Kimi K2.5 measured
IFBench70.2%26.4%Kimi K2.5
Long context reasoning (AA-LCR)78.0%0.0%Kimi K2.5
Fiction.LiveBench86.1%Not measuredOnly Kimi K2.5 measured
CL-bench19.3%Not measuredOnly Kimi K2.5 measured
CL-bench Life13.2%Not measuredOnly Kimi K2.5 measured
LMArena Text14451234Kimi K2.5
LMArena Text: instruction following14311214Kimi K2.5
LMArena Text: creative writing14221194Kimi K2.5
LMArena Vision1269Not measuredOnly Kimi K2.5 measured
LMArena Text: English14531256Kimi K2.5
LMArena Text: non-English14331198Kimi K2.5
LMArena Text: Chinese14941204Kimi K2.5
LMArena Text: German14411211Kimi K2.5
LMArena Text: Japanese14231111Kimi K2.5
LMArena Text: Korean14121188Kimi K2.5
LMArena Text: French14541204Kimi K2.5
LMArena Text: Spanish1450Not measuredOnly Kimi K2.5 measured
LMArena Text: Russian14351216Kimi K2.5
LMArena Text: Polish1445Not measuredOnly Kimi K2.5 measured
LMArena Text: multi-turn14441217Kimi K2.5
LMArena Text: longer queries14451246Kimi K2.5
LMArena Text: writing and language14251214Kimi K2.5
LMArena Text: entertainment and media14171184Kimi K2.5
LMArena Vision: English1274Not measuredOnly Kimi K2.5 measured
LMArena Vision: Chinese1328Not measuredOnly Kimi K2.5 measured
LMArena Vision: reading text1276Not measuredOnly Kimi K2.5 measured
LMArena Vision: humor1284Not measuredOnly Kimi K2.5 measured
LMArena Vision: captioning1257Not measuredOnly Kimi K2.5 measured
LMArena Vision: recognizing things1285Not measuredOnly Kimi K2.5 measured
LMArena Vision: creative writing1269Not measuredOnly Kimi K2.5 measured
Neither model has a result on the other 14 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Kimi K2.5 and Mistral Small 3
BenchmarkKimi K2.5Mistral Small 3Leader
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

Kimi K2.5$0.0020Fits in one request
Mistral Small 3$0.0001Fits in one request

Mistral Small 3 costs less

Repository review

50K input + 3K output tokens

Kimi K2.5$0.035Fits in one request
Mistral Small 3$0.0027Exceeds the context window

Mistral Small 3 costs less

Agent loop

200K input + 10K output tokens

Kimi K2.5$0.135Fits in one request
Mistral Small 3$0.011Exceeds the context window

Mistral Small 3 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 Kimi K2.5 and Mistral Small 3
SpecificationMoonshotAI Kimi K2.5Mistral Small 3
MakerMoonshot AIMistral AI
WeightsOpen weightsOpen weights
ReleasedJanuary 27, 2026January 30, 2025
Context window262K33K
Input $/Mtok$0.532$0.05
Output $/Mtok$2.85$0.08
Coding index38.6Not benchmarked
Independent measurements behind it30
Cost per solved task$0.24Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Kimi K2.5 better than Mistral Small 3 for coding?

Only Kimi K2.5 has a coding score, 38.6 on the Vibe Coding Index; Mistral Small 3 has none, so the two cannot be ranked on coding. Pick Mistral Small 3 for output tokens 35.6x cheaper ($0.08 vs $2.85 per million).

Which is cheaper, Kimi K2.5 or Mistral Small 3?

Kimi K2.5 costs $0.532 per million input tokens and $2.85 per million output tokens; Mistral Small 3 costs $0.05 and $0.08. Cost per solved task is known for Kimi K2.5 only.

Which has the bigger context window, Kimi K2.5 or Mistral Small 3?

Kimi K2.5, with 262K tokens against 33K.

How many benchmarks is the Kimi K2.5 vs Mistral Small 3 comparison based on?

32 benchmarks have a result for both models as of October 21, 2026. Kimi K2.5's coding index rests on 3 independent benchmarks and Mistral Small 3's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next