Enter opens the highlighted result. Esc closes.

GPT-5.3-Codex vs Kimi K2 Thinking

GPT-5.3-Codex leads Kimi K2 Thinking on the Artificial Analysis Intelligence Index, 32.5 to 22.0, and Kimi K2 Thinking costs 5.6x less per output token ($2.50 vs $14.00 per million). Updated October 22, 2026; 15 benchmarks have a result for both.

Replace GPT-5.3-Codex Replace Kimi K2 Thinking
GPT-5.3-CodexOpenAI
Coding
60.0
Overall
No data
Output $/M
$14.00
14
Kimi K2 ThinkingMoonshot AI
Coding
No data
Overall
No data
Output $/M
$2.50
1
vs boards won of 15 shared

GPT-5.3-Codex or Kimi K2 Thinking: which to pick

Only GPT-5.3-Codex has a coding score, 60.0 on the Vibe Coding Index; Kimi K2 Thinking has none, so the two cannot be ranked on coding.

Pick Kimi K2 Thinking for output tokens 5.6x cheaper ($2.50 vs $14.00 per million) and work measured by τ²-Bench Telecom, where it leads.

Model AGPT-5.3-CodexOpenAI60.0/1003 independent benchmarks$0.59 per solved taskModel BKimi K2 ThinkingMoonshot AINot benchmarked1 independent benchmark$0.61 per solved task
Shared results
15
GPT-5.3-Codex only
19
Kimi K2 Thinking only
15
Like-for-like categories
6 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
GPT-5.3-Codex 60.0 from 3 benchmarks, Kimi K2 Thinking not benchmarked.
Widest category
Coding, 29.7 points apart on average over 6 shared benchmarks.
Widest single gap
Terminal-Bench: GPT-5.3-Codex 78.4% against Kimi K2 Thinking 35.7%, 42.7 points apart on a 0 to 100 scale.
Output price
GPT-5.3-Codex $14.00 against Kimi K2 Thinking $2.50 per million output tokens, 5.6x apart.
Context window
GPT-5.3-Codex takes 400K tokens, Kimi K2 Thinking 262K.

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
CategoryGPT-5.3-CodexBarsKimi K2 ThinkingTies
Coding60
Intelligence20
Agentic work11
Reasoning20
Web design10
General20
All shared boards141
CodingLike-for-likeSix of these boards feed the Vibe Coding Index60
Coding benchmarks for GPT-5.3-Codex and Kimi K2 Thinking
BenchmarkGPT-5.3-CodexKimi K2 ThinkingLeader
SWE-bench VerifiedNot measured63.4%Only Kimi K2 Thinking measured
LMArena WebDev1407Not measuredOnly GPT-5.3-Codex measured
LiveCodeBench (Vals AI)87.3%63.1%GPT-5.3-Codex
SWE-bench Verified (Vals AI)78.0%60.2%GPT-5.3-Codex
Vibe Code Bench (Vals AI)61.8%Not measuredOnly GPT-5.3-Codex measured
IOI (Vals AI)53.8%Not measuredOnly GPT-5.3-Codex measured
LiveCodeBench (Artificial Analysis run)Not measured85.3%Only Kimi K2 Thinking measured
Terminal-Bench Hard (Artificial Analysis run)53.0%31.1%GPT-5.3-Codex
Terminal-Bench78.4%35.7%GPT-5.3-Codex
WeirdML79.3%42.8%GPT-5.3-Codex
ALE-Bench1655.2597.5GPT-5.3-Codex
LMArena WebDev: React1395Not measuredOnly GPT-5.3-Codex measured
LMArena WebDev: HTML1422Not measuredOnly GPT-5.3-Codex measured
LMArena WebDev: from an image1443Not measuredOnly GPT-5.3-Codex measured
Neither model has a result on the other 25 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GPT-5.3-Codex and Kimi K2 Thinking
BenchmarkGPT-5.3-CodexKimi K2 ThinkingLeader
AA Intelligence Index32.522.0GPT-5.3-Codex
Epoch Capabilities Index157.1145.8GPT-5.3-Codex
Agentic workLike-for-likePublic boards, shown as published, kept out of the index11
Agentic work benchmarks for GPT-5.3-Codex and Kimi K2 Thinking
BenchmarkGPT-5.3-CodexKimi K2 ThinkingLeader
Design Arena: agents building web apps1032Not measuredOnly GPT-5.3-Codex measured
Design Arena: agents building full-stack apps983Not measuredOnly GPT-5.3-Codex measured
Design Arena: agents building mobile apps1053Not measuredOnly GPT-5.3-Codex measured
Design Arena: agents building Android apps1084Not measuredOnly GPT-5.3-Codex measured
Design Arena: agents building Godot games1124Not measuredOnly GPT-5.3-Codex measured
τ²-Bench Telecom86.0%93.0%Kimi K2 Thinking
METR time horizons74.5%59.2%GPT-5.3-Codex
Vending-Bench 2$5,940Not measuredOnly GPT-5.3-Codex measured
Neither model has a result on the other 33 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index20
Reasoning benchmarks for GPT-5.3-Codex and Kimi K2 Thinking
BenchmarkGPT-5.3-CodexKimi K2 ThinkingLeader
GPQA DiamondNot measured78.5%Only Kimi K2 Thinking measured
GPQA Diamond (Artificial Analysis run)91.5%83.8%GPT-5.3-Codex
Humanity's Last Exam (Artificial Analysis run)42.5%23.8%GPT-5.3-Codex
Neither model has a result on the other 19 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
Math benchmarks for GPT-5.3-Codex and Kimi K2 Thinking
BenchmarkGPT-5.3-CodexKimi K2 ThinkingLeader
AIMENot measured85.4%Only Kimi K2 Thinking measured
MGSM (Vals AI)Not measured90.2%Only Kimi K2 Thinking measured
AA Math IndexNot measured94.7Only Kimi K2 Thinking measured
AIME 2025 (Artificial Analysis run)Not measured94.7%Only Kimi K2 Thinking measured
Neither model has a result on the other 9 boards here
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Knowledge benchmarks for GPT-5.3-Codex and Kimi K2 Thinking
BenchmarkGPT-5.3-CodexKimi K2 ThinkingLeader
MMLU ProNot measured81.1%Only Kimi K2 Thinking measured
LegalBench (Vals AI)Not measured80.2%Only Kimi K2 Thinking measured
CaseLaw (Vals AI)Not measured65.7%Only Kimi K2 Thinking measured
CorpFin (Vals AI)Not measured60.6%Only Kimi K2 Thinking measured
TaxEval (Vals AI)Not measured71.7%Only Kimi K2 Thinking measured
MedQA (Vals AI)Not measured92.6%Only Kimi K2 Thinking measured
MMLU Pro (Artificial Analysis run)Not measured84.8%Only Kimi K2 Thinking measured
Neither model has a result on the other 11 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index10
Web design benchmarks for GPT-5.3-Codex and Kimi K2 Thinking
BenchmarkGPT-5.3-CodexKimi K2 ThinkingLeader
Design Arena: websites11621113GPT-5.3-Codex
Design Arena: UI components1134Not measuredOnly GPT-5.3-Codex measured
Design Arena: game dev1161Not measuredOnly GPT-5.3-Codex measured
Design Arena: data viz1164Not measuredOnly GPT-5.3-Codex measured
Design Arena: 3D1007Not measuredOnly GPT-5.3-Codex measured
Design Arena: all code categories1144Not measuredOnly GPT-5.3-Codex measured
Design Arena: SVG1129Not measuredOnly GPT-5.3-Codex measured
Design Arena: ASCII art1152Not measuredOnly GPT-5.3-Codex measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index20
General benchmarks for GPT-5.3-Codex and Kimi K2 Thinking
BenchmarkGPT-5.3-CodexKimi K2 ThinkingLeader
IFBench75.4%68.1%GPT-5.3-Codex
Long context reasoning (AA-LCR)83.3%72.0%GPT-5.3-Codex
CL-benchNot measured17.6%Only Kimi K2 Thinking measured
Neither model has a result on the other 43 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-5.3-Codex and Kimi K2 Thinking
BenchmarkGPT-5.3-CodexKimi K2 ThinkingLeader
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

GPT-5.3-Codex$0.0088Fits in one request
Kimi K2 Thinking$0.0019Fits in one request

Kimi K2 Thinking costs less

Repository review

50K input + 3K output tokens

GPT-5.3-Codex$0.130Fits in one request
Kimi K2 Thinking$0.037Fits in one request

Kimi K2 Thinking costs less

Agent loop

200K input + 10K output tokens

GPT-5.3-Codex$0.490Fits in one request
Kimi K2 Thinking$0.145Fits in one request

Kimi K2 Thinking 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 GPT-5.3-Codex and Kimi K2 Thinking
SpecificationOpenAI GPT-5.3-CodexMoonshotAI Kimi K2 Thinking
MakerOpenAIMoonshot AI
WeightsNo public weights listedOpen weights
ReleasedFebruary 24, 2026November 6, 2025
Context window400K262K
Input $/Mtok$1.75$0.60
Output $/Mtok$14.00$2.50
Coding index60.0Not benchmarked
Independent measurements behind it31
Cost per solved task$0.59$0.61
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5.3-Codex better than Kimi K2 Thinking for coding?

Only GPT-5.3-Codex has a coding score, 60.0 on the Vibe Coding Index; Kimi K2 Thinking has none, so the two cannot be ranked on coding. Pick Kimi K2 Thinking for output tokens 5.6x cheaper ($2.50 vs $14.00 per million) and work measured by τ²-Bench Telecom, where it leads.

Which is cheaper, GPT-5.3-Codex or Kimi K2 Thinking?

GPT-5.3-Codex costs $1.75 per million input tokens and $14.00 per million output tokens; Kimi K2 Thinking costs $0.60 and $2.50. Per solved SWE-bench Verified task, GPT-5.3-Codex costs $0.59 and Kimi K2 Thinking $0.61.

Which has the bigger context window, GPT-5.3-Codex or Kimi K2 Thinking?

GPT-5.3-Codex, with 400K tokens against 262K.

How many benchmarks is the GPT-5.3-Codex vs Kimi K2 Thinking comparison based on?

15 benchmarks have a result for both models as of October 22, 2026. GPT-5.3-Codex's coding index rests on 3 independent benchmarks and Kimi K2 Thinking's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next