Enter opens the highlighted result. Esc closes.

R1 0528 vs Kimi K2 0905

Kimi K2 0905 leads R1 0528 on the Artificial Analysis Intelligence Index, 15.3 to 13.1, at about the same output price ($2.50 vs $2.18 per million). Updated October 22, 2026; 16 benchmarks have a result for both.

Replace R1 0528 Replace Kimi K2 0905
R1 0528DeepSeek
Coding
No data
Overall
46.8
Output $/M
$2.18
12
Kimi K2 0905Moonshot AI
Coding
No data
Overall
No data
Output $/M
$2.50
4
vs boards won of 16 shared

R1 0528 or Kimi K2 0905: which to pick

Neither model has a coding score yet, so the two cannot be ranked on coding.

Pick R1 0528 for output tokens 1.1x cheaper ($2.18 vs $2.50 per million) and work measured by Terminal-Bench Hard (Artificial Analysis run) or AA Intelligence Index, where it leads.

Model AR1 0528DeepSeekNot benchmarkedNo coding benchmarkNo cost per solved taskModel BKimi K2 0905Moonshot AINot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
16
R1 0528 only
44
Kimi K2 0905 only
1
Like-for-like categories
8 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
R1 0528 not benchmarked, Kimi K2 0905 not benchmarked.
Widest category
Agentic work, 36.8 points apart on average over 1 shared benchmark.
Widest single gap
τ²-Bench Telecom: R1 0528 36.5% against Kimi K2 0905 73.4%, 36.8 points apart on a 0 to 100 scale.
Output price
R1 0528 $2.18 against Kimi K2 0905 $2.50 per million output tokens, 1.1x apart.
Context window
R1 0528 takes 164K tokens, Kimi K2 0905 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
CategoryR1 0528BarsKimi K2 0905Ties
Coding31
Intelligence01
Agentic work01
Reasoning20
Math20
Knowledge10
Web design20
General21
All shared boards124
CodingLike-for-likeSix of these boards feed the Vibe Coding Index31
Coding benchmarks for R1 0528 and Kimi K2 0905
BenchmarkR1 0528Kimi K2 0905Leader
Aider polyglot71.4%Not measuredOnly R1 0528 measured
LiveCodeBench (Artificial Analysis run)77.0%61.0%R1 0528
Terminal-Bench Hard (Artificial Analysis run)15.9%23.5%Kimi K2 0905
WeirdML41.6%36.7%R1 0528
ALE-Bench804.1267.1R1 0528
LMArena Text: coding1427Not measuredOnly R1 0528 measured
LMArena Text: software and IT1435Not measuredOnly R1 0528 measured
Neither model has a result on the other 32 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for R1 0528 and Kimi K2 0905
BenchmarkR1 0528Kimi K2 0905Leader
AA Intelligence Index13.115.3Kimi K2 0905
Neither model has a result on the other 1 board here
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for R1 0528 and Kimi K2 0905
BenchmarkR1 0528Kimi K2 0905Leader
τ²-Bench Telecom36.5%73.4%Kimi K2 0905
METR time horizons53.8%Not measuredOnly R1 0528 measured
DeepResearch Bench35.1%Not measuredOnly R1 0528 measured
BrowseComp-Plus16.4%Not measuredOnly R1 0528 measured
Neither model has a result on the other 37 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index20
Reasoning benchmarks for R1 0528 and Kimi K2 0905
BenchmarkR1 0528Kimi K2 0905Leader
GPQA Diamond (Artificial Analysis run)81.3%76.7%R1 0528
Humanity's Last Exam (Artificial Analysis run)15.8%6.4%R1 0528
GPQA Diamond (Epoch AI run)76.3%Not measuredOnly R1 0528 measured
ARC-AGI21.2%Not measuredOnly R1 0528 measured
ARC-AGI-21.1%Not measuredOnly R1 0528 measured
SimpleBench40.8%Not measuredOnly R1 0528 measured
LMArena Text: hard prompts1415Not measuredOnly R1 0528 measured
LMArena Text: expert1394Not measuredOnly R1 0528 measured
LMArena Text: hard prompts in English1422Not measuredOnly R1 0528 measured
Kagi LLM BenchmarkNot measured52.2%Only Kimi K2 0905 measured
Neither model has a result on the other 12 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index20
Math benchmarks for R1 0528 and Kimi K2 0905
BenchmarkR1 0528Kimi K2 0905Leader
AA Math Index76.057.3R1 0528
AIME 2025 (Artificial Analysis run)76.0%57.3%R1 0528
OTIS Mock AIME66.4%Not measuredOnly R1 0528 measured
MATH level 596.6%Not measuredOnly R1 0528 measured
LMArena Text: math1400Not measuredOnly R1 0528 measured
LMArena Text: mathematical work1415Not measuredOnly R1 0528 measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index10
Knowledge benchmarks for R1 0528 and Kimi K2 0905
BenchmarkR1 0528Kimi K2 0905Leader
MMLU Pro (Artificial Analysis run)84.9%81.9%R1 0528
LMArena Text: science1430Not measuredOnly R1 0528 measured
LMArena Text: medicine and healthcare1445Not measuredOnly R1 0528 measured
LMArena Text: legal and government1419Not measuredOnly R1 0528 measured
LMArena Text: business and finance1402Not measuredOnly R1 0528 measured
Neither model has a result on the other 13 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index20
Web design benchmarks for R1 0528 and Kimi K2 0905
BenchmarkR1 0528Kimi K2 0905Leader
Design Arena: websites11491088R1 0528
Design Arena: UI components1100Not measuredOnly R1 0528 measured
Design Arena: game dev1097Not measuredOnly R1 0528 measured
Design Arena: data viz1182Not measuredOnly R1 0528 measured
Design Arena: 3D1110Not measuredOnly R1 0528 measured
Design Arena: all code categories11381098R1 0528
Design Arena: SVG1038Not measuredOnly R1 0528 measured
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index21
General benchmarks for R1 0528 and Kimi K2 0905
BenchmarkR1 0528Kimi K2 0905Leader
IFBench39.6%41.7%Kimi K2 0905
Long context reasoning (AA-LCR)55.7%53.0%R1 0528
Fiction.LiveBench75.0%66.7%R1 0528
Creative writing (Lech Mazur)81.9%Not measuredOnly R1 0528 measured
LMArena Text1428Not measuredOnly R1 0528 measured
LMArena Text: instruction following1382Not measuredOnly R1 0528 measured
LMArena Text: creative writing1405Not measuredOnly R1 0528 measured
LMArena Text: English1437Not measuredOnly R1 0528 measured
LMArena Text: non-English1412Not measuredOnly R1 0528 measured
LMArena Text: Chinese1442Not measuredOnly R1 0528 measured
LMArena Text: German1404Not measuredOnly R1 0528 measured
LMArena Text: Japanese1391Not measuredOnly R1 0528 measured
LMArena Text: Korean1360Not measuredOnly R1 0528 measured
LMArena Text: French1417Not measuredOnly R1 0528 measured
LMArena Text: Spanish1411Not measuredOnly R1 0528 measured
LMArena Text: Russian1423Not measuredOnly R1 0528 measured
LMArena Text: Polish1428Not measuredOnly R1 0528 measured
LMArena Text: multi-turn1405Not measuredOnly R1 0528 measured
LMArena Text: longer queries1391Not measuredOnly R1 0528 measured
LMArena Text: writing and language1402Not measuredOnly R1 0528 measured
LMArena Text: entertainment and media1400Not measuredOnly R1 0528 measured
Neither model has a result on the other 25 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for R1 0528 and Kimi K2 0905
BenchmarkR1 0528Kimi K2 0905Leader
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

R1 0528$0.0016Fits in one request
Kimi K2 0905$0.0019Fits in one request

R1 0528 costs less

Repository review

50K input + 3K output tokens

R1 0528$0.032Fits in one request
Kimi K2 0905$0.037Fits in one request

R1 0528 costs less

Agent loop

200K input + 10K output tokens

R1 0528$0.122Exceeds the context window
Kimi K2 0905$0.145Fits in one request

R1 0528 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 R1 0528 and Kimi K2 0905
SpecificationDeepSeek R1 0528MoonshotAI Kimi K2 0905
MakerDeepSeekMoonshot AI
WeightsOpen weightsOpen weights
ReleasedMay 28, 2025September 4, 2025
Context window164K262K
Input $/Mtok$0.50$0.60
Output $/Mtok$2.18$2.50
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is R1 0528 better than Kimi K2 0905 for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick R1 0528 for output tokens 1.1x cheaper ($2.18 vs $2.50 per million) and work measured by Terminal-Bench Hard (Artificial Analysis run) or AA Intelligence Index, where it leads.

Which is cheaper, R1 0528 or Kimi K2 0905?

R1 0528 costs $0.50 per million input tokens and $2.18 per million output tokens; Kimi K2 0905 costs $0.60 and $2.50. Cost per solved task is known for neither of them.

Which has the bigger context window, R1 0528 or Kimi K2 0905?

Kimi K2 0905, with 262K tokens against 164K.

How many benchmarks is the R1 0528 vs Kimi K2 0905 comparison based on?

16 benchmarks have a result for both models as of October 22, 2026. R1 0528's coding index rests on 0 independent benchmarks and Kimi K2 0905's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next