Noometry

Model comparison

Kimi K2.5 vs Qwen3.7 Max

Qwen3.7 Max is the stronger model overall, scoring 51.5 to 48.1 on the Noometry Index. Kimi K2.5 costs 4.2× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.

Last verified . 24 shared benchmarks.

Kimi K2.5 Moonshot AI

48.1

Rank #57 Confirmed

Qwen3.7 Max Alibaba (Qwen)

51.5

Rank #42 Confirmed

Summary

  • They share 24 benchmarks with published results for both. Kimi K2.5 scores higher in 3 categories and Qwen3.7 Max in 6 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Qwen3.7 Max leads 49.2 to 31.2.
  • The biggest single-benchmark swing is SimpleBench: 46.8% for Kimi K2.5 and 70.4% for Qwen3.7 Max.
  • Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $2.50 / $7.50 for Qwen3.7 Max.
  • Qwen3.7 Max accepts more context: 1M tokens versus 262K.
  • Kimi K2.5 has downloadable open weights; the other is API-only.

Side by side

Kimi K2.5Qwen3.7 Max
ProviderMoonshot AIAlibaba (Qwen)
Noometry Index48.151.5
Released2026-01-272026-05-19
WeightsOpenProprietary
Context window262K1M
Max output262K131K
Input $ / M tokens$0.45$2.50
Output $ / M tokens$2.25$7.50
Results tracked5133

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

Category by category

Coding Qwen3.7 Max leads

Kimi K2.5: 48.8 (#53), Qwen3.7 Max: 50.4 (#45)

BenchmarkKimi K2.5Qwen3.7 Max
SWE-bench Verified73.8%77.3%
LMArena WebDev14371515
SciCode49%48.8%
LMArena Coding14741498
ALE-Bench821.651,189
SWE-bench Verified (bash only)70.8%—
SWE-bench Multilingual67.3%—
WeirdML45.6%—

Agentic & Tool Use Kimi K2.5 leads

Kimi K2.5: 34.2 (#48), Qwen3.7 Max: 22.1 (#135)

BenchmarkKimi K2.5Qwen3.7 Max
Terminal-Bench43.2%—
OSWorld63.3%—
GBAEval—0.4%
Vending-Bench 21,198—

Reasoning Qwen3.7 Max leads

Kimi K2.5: 31.2 (#80), Qwen3.7 Max: 49.2 (#38)

BenchmarkKimi K2.5Qwen3.7 Max
SimpleBench46.8%70.4%
NYT Connections (extended)69.9%85.1%
CritPt3.1%13.4%
Chess Puzzles12%19%
LMArena Hard Prompts14531483
Epoch Capabilities Index148.03153.68
ARC-AGI-211.8%—
Kagi LLM Benchmark78.5%—
ARC-AGI-165.3%—
EnigmaEval3.4%—
Thematic Generalization69.4%—
EBR-Bench—9.5%
Mystery Game Puzzles—32%
DTBench—92.3%
LMCA—44%

Math Qwen3.7 Max leads

Kimi K2.5: 51.8 (#53), Qwen3.7 Max: 62.4 (#32)

BenchmarkKimi K2.5Qwen3.7 Max
OTIS Mock AIME 2024-202592.2%95.6%
LMArena Math14701490
FrontierMath (Tiers 1-3)—64.6%
FrontierMath Tier 4—34.1%
MathArena Final-Answer Competitions62.3%—
ProofBench—26%
FrontierMath (Feb 2025 set)27.9%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge Qwen3.7 Max leads

Kimi K2.5: 53.6 (#56), Qwen3.7 Max: 61.6 (#28)

BenchmarkKimi K2.5Qwen3.7 Max
GPQA Diamond87.6%90.9%
SimpleQA Verified34.3%55.8%
LMArena Expert14661488
Humanity's Last Exam24.4%—
Vectara Hallucination Rate14.2%—

Multimodal Not comparable

Kimi K2.5: 41.1 (#39), Qwen3.7 Max: —

BenchmarkKimi K2.5Qwen3.7 Max
LMArena Vision1269—
LMArena Document1430—

Multilingual Qwen3.7 Max leads

Kimi K2.5: 53.9 (#53), Qwen3.7 Max: 56.9 (#15)

BenchmarkKimi K2.5Qwen3.7 Max
LMArena Non-English14331474
LMArena Chinese14951530
LMArena Russian14351484
LMArena French1454—
LMArena German1441—
LMArena Japanese1421—
LMArena Korean1410—
LMArena Spanish1450—

Instruction Following Qwen3.7 Max leads

Kimi K2.5: 75.3 (#64), Qwen3.7 Max: 76.7 (#38)

BenchmarkKimi K2.5Qwen3.7 Max
LMArena Instruction Following14311460

Long Context Kimi K2.5 leads

Kimi K2.5: 52.1 (#7), Qwen3.7 Max: 45.4 (#40)

BenchmarkKimi K2.5Qwen3.7 Max
LMArena Longer Query14451482
Fiction.LiveBench86.1%—
CL-bench19.3%—
CL-bench Life13.2%—

Writing & Preference Too close to call

Kimi K2.5: 65.1 (#53), Qwen3.7 Max: 65.0 (#54)

BenchmarkKimi K2.5Qwen3.7 Max
LMArena Text14451476
LMArena Creative Writing14231449
LMArena Multi-Turn14441481
EQ-Bench Creative Writing1579—
EQ-Bench 4—1110

Frequently asked questions

Is Kimi K2.5 better than Qwen3.7 Max?

Qwen3.7 Max is the stronger model overall, scoring 51.5 to 48.1 on the Noometry Index. Kimi K2.5 costs 4.2× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.

Which is cheaper, Kimi K2.5 or Qwen3.7 Max?

Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; Qwen3.7 Max lists at $2.50 and $7.50.

Is Kimi K2.5 or Qwen3.7 Max better for coding?

Qwen3.7 Max scores higher on coding benchmarks: 50.4 versus 48.8 in the Noometry coding category.

Which has the bigger context window?

Qwen3.7 Max does, with 1M tokens against 262K.

How many benchmarks do Kimi K2.5 and Qwen3.7 Max share?

24 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and Qwen3.7 Max has 33.

Related comparisons

Go deeper