Noometry

Model comparison

Claude Opus 4.8 vs Kimi K2.5

Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 47.8 on the Noometry Index. Kimi K2.5 costs 11× less per token, which makes it the better buy when Claude Opus 4.8's lead doesn't matter for your workload.

Last verified . 40 shared benchmarks.

Claude Opus 4.8 Anthropic

60.7

Rank #9 Confirmed

Kimi K2.5 Moonshot AI

47.8

Rank #50 Confirmed

Summary

  • They share 40 benchmarks with published results for both. Claude Opus 4.8 scores higher in 9 categories and Kimi K2.5 in 1 category; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Claude Opus 4.8 leads 64.6 to 31.2.
  • The biggest single-benchmark swing is ARC-AGI-2: 72.1% for Claude Opus 4.8 and 11.8% for Kimi K2.5.
  • Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $5 / $25 for Claude Opus 4.8.
  • Claude Opus 4.8 accepts more context: 1M tokens versus 262K.
  • Kimi K2.5 has downloadable open weights; the other is API-only.

Side by side

Claude Opus 4.8Kimi K2.5
ProviderAnthropicMoonshot AI
Noometry Index60.747.8
Released2026-05-282026-01-27
WeightsProprietaryOpen
Context window1M262K
Max output128K262K
Input $ / M tokens$5$0.45
Output $ / M tokens$25$2.25
Results tracked6551

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Claude Opus 4.8 leads

Claude Opus 4.8: 60.2 (#9), Kimi K2.5: 48.7 (#48)

BenchmarkClaude Opus 4.8Kimi K2.5
LMArena WebDev15561436
SciCode53.5%49%
WeirdML82.9%45.6%
LMArena Coding14911474
ALE-Bench1,564821.65
SWE-bench Verified—73.8%
DeepSWE59%—
FrontierCode46.5%—
SWE-bench Verified (bash only)—70.8%
SWE-bench Multilingual—67.3%
GSO47.1%—

Agentic & Tool Use Claude Opus 4.8 leads

Claude Opus 4.8: 47.6 (#10), Kimi K2.5: 34.0 (#44)

BenchmarkClaude Opus 4.8Kimi K2.5
Vending-Bench 25,7871,198
Terminal-Bench—43.2%
APEX-Agents48.9%—
OSWorld 2.020.6%—
Remote Labor Index8.3%—
τ²-bench Banking39.7%—
DeepResearch Bench50.2%—
OSWorld—63.3%
PostTrainBench33.8%—
GBAEval70.9%—
GDP.pdf24%—
LMArena Search1204—

Reasoning Claude Opus 4.8 leads

Claude Opus 4.8: 64.6 (#14), Kimi K2.5: 31.2 (#74)

BenchmarkClaude Opus 4.8Kimi K2.5
ARC-AGI-272.1%11.8%
SimpleBench64.8%46.8%
Kagi LLM Benchmark88.8%78.5%
NYT Connections (extended)91.1%69.9%
ARC-AGI-192.5%65.3%
CritPt20.9%3.1%
Chess Puzzles34%12%
EnigmaEval23.5%3.4%
LMArena Hard Prompts14811454
Epoch Capabilities Index158.3148.05
Thematic Generalization—69.4%
EBR-Bench28.6%—
Mystery Game Puzzles36%—
DTBench94.9%—
LMCA57.5%—
Surface Evolver Bench87.5%—
Bench to the Future 30.14—
ForecastBench59.9—

Math Claude Opus 4.8 leads

Claude Opus 4.8: 78.2 (#11), Kimi K2.5: 50.3 (#46)

BenchmarkClaude Opus 4.8Kimi K2.5
MathArena Final-Answer Competitions91.8%62.3%
OTIS Mock AIME 2024-202598.3%92.2%
LMArena Math14871470
FrontierMath (Feb 2025 set)47.2%27.9%
FrontierMath Tier 4 (v1)31.3%4.2%
FrontierMath (Tiers 1-3)80%—
FrontierMath Tier 456.1%—
ProofBench69%—

Knowledge Claude Opus 4.8 leads

Claude Opus 4.8: 61.3 (#26), Kimi K2.5: 53.6 (#52)

BenchmarkClaude Opus 4.8Kimi K2.5
GPQA Diamond91%87.6%
SimpleQA Verified53%34.3%
LMArena Expert15011466
Humanity's Last Exam—24.4%
Vectara Hallucination Rate—14.2%

Multimodal Claude Opus 4.8 leads

Claude Opus 4.8: 43.2 (#20), Kimi K2.5: 40.4 (#37)

BenchmarkClaude Opus 4.8Kimi K2.5
LMArena Vision12951269
LMArena Document14751430
Blueprint-Bench 214.5%—
Furniture Assembly42.5%—

Multilingual Claude Opus 4.8 leads

Claude Opus 4.8: 55.4 (#31), Kimi K2.5: 54.1 (#50)

BenchmarkClaude Opus 4.8Kimi K2.5
LMArena Non-English14501433
LMArena Chinese15091495
LMArena French14821454
LMArena German14701441
LMArena Japanese14401422
LMArena Korean14321412
LMArena Russian14761436
LMArena Spanish14701450

Instruction Following Claude Opus 4.8 leads

Claude Opus 4.8: 77.6 (#22), Kimi K2.5: 75.5 (#60)

BenchmarkClaude Opus 4.8Kimi K2.5
LMArena Instruction Following14761431

Long Context Kimi K2.5 leads

Claude Opus 4.8: 45.4 (#37), Kimi K2.5: 52.1 (#7)

BenchmarkClaude Opus 4.8Kimi K2.5
LMArena Longer Query14821445
Fiction.LiveBench—86.1%
CL-bench—19.3%
CL-bench Life—13.2%

Writing & Preference Claude Opus 4.8 leads

Claude Opus 4.8: 72.1 (#17), Kimi K2.5: 65.3 (#51)

BenchmarkClaude Opus 4.8Kimi K2.5
LMArena Text14601445
LMArena Creative Writing14531423
EQ-Bench Creative Writing18401579
LMArena Multi-Turn14761444
EQ-Bench 41281—

Frequently asked questions

Is Claude Opus 4.8 better than Kimi K2.5?

Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 47.8 on the Noometry Index. Kimi K2.5 costs 11× less per token, which makes it the better buy when Claude Opus 4.8's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 4.8 or Kimi K2.5?

Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; Claude Opus 4.8 lists at $5 and $25.

Is Claude Opus 4.8 or Kimi K2.5 better for coding?

Claude Opus 4.8 scores higher on coding benchmarks: 60.2 versus 48.7 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 4.8 does, with 1M tokens against 262K.

How many benchmarks do Claude Opus 4.8 and Kimi K2.5 share?

40 benchmarks have published results for both models. Claude Opus 4.8 has 65 scored results on Noometry and Kimi K2.5 has 51.

Related comparisons

Go deeper