Model comparison
Kimi K2.5 vs Qwen3.7 Plus
Kimi K2.5 is the stronger model overall, scoring 48.1 to 45.3 on the Noometry Index.
Last verified . 26 shared benchmarks.
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. Kimi K2.5 scores higher in 5 categories and Qwen3.7 Plus in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Kimi K2.5 leads 34.2 to 21.4.
- The biggest single-benchmark swing is Chess Puzzles: 12% for Kimi K2.5 and 24% for Qwen3.7 Plus.
- Qwen3.7 Plus is cheaper at $0.40 / $1.60 per million input/output tokens, against $0.45 / $2.25 for Kimi K2.5.
- Qwen3.7 Plus accepts more context: 1M tokens versus 262K.
- Kimi K2.5 has downloadable open weights; the other is API-only.
Side by side
| Kimi K2.5 | Qwen3.7 Plus | |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 48.1 | 45.3 |
| Released | 2026-01-27 | 2026-06-02 |
| Weights | Open | Proprietary |
| Context window | 262K | 1M |
| Max output | 262K | 131K |
| Input $ / M tokens | $0.45 | $0.40 |
| Output $ / M tokens | $2.25 | $1.60 |
| Results tracked | 51 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Kimi K2.5: 48.8 (#53), Qwen3.7 Plus: 36.6 (#206)
| Benchmark | Kimi K2.5 | Qwen3.7 Plus |
|---|---|---|
| SciCode | 49% | 45.5% |
| LMArena Coding | 1474 | 1473 |
| SWE-bench Verified | 73.8% | — |
| FrontierCode | — | 10.2% |
| SWE-bench Verified (bash only) | 70.8% | — |
| LMArena WebDev | 1437 | — |
| SWE-bench Multilingual | 67.3% | — |
| WeirdML | 45.6% | — |
| ALE-Bench | 821.65 | — |
Agentic & Tool Use Kimi K2.5 leads
Kimi K2.5: 34.2 (#48), Qwen3.7 Plus: 21.4 (#138)
| Benchmark | Kimi K2.5 | Qwen3.7 Plus |
|---|---|---|
| Terminal-Bench | 43.2% | — |
| OSWorld 2.0 | — | 2.8% |
| OSWorld | 63.3% | — |
| Vending-Bench 2 | 1,198 | — |
Reasoning Qwen3.7 Plus leads
Kimi K2.5: 31.2 (#80), Qwen3.7 Plus: 39.3 (#59)
| Benchmark | Kimi K2.5 | Qwen3.7 Plus |
|---|---|---|
| NYT Connections (extended) | 69.9% | 74.8% |
| CritPt | 3.1% | 9.1% |
| Chess Puzzles | 12% | 24% |
| LMArena Hard Prompts | 1453 | 1460 |
| Epoch Capabilities Index | 148.03 | 147.37 |
| ARC-AGI-2 | 11.8% | — |
| SimpleBench | 46.8% | — |
| Kagi LLM Benchmark | 78.5% | — |
| ARC-AGI-1 | 65.3% | — |
| EnigmaEval | 3.4% | — |
| Thematic Generalization | 69.4% | — |
| Mystery Game Puzzles | — | 17% |
| DTBench | — | 84% |
| LMCA | — | 37.6% |
Math Kimi K2.5 leads
Kimi K2.5: 51.8 (#53), Qwen3.7 Plus: 50.5 (#56)
| Benchmark | Kimi K2.5 | Qwen3.7 Plus |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 93.3% |
| LMArena Math | 1470 | 1466 |
| FrontierMath (Tiers 1-3) | — | 34.4% |
| MathArena Final-Answer Competitions | 62.3% | — |
| FrontierMath (Feb 2025 set) | 27.9% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Qwen3.7 Plus leads
Kimi K2.5: 53.6 (#56), Qwen3.7 Plus: 54.9 (#51)
| Benchmark | Kimi K2.5 | Qwen3.7 Plus |
|---|---|---|
| GPQA Diamond | 87.6% | 87.9% |
| LMArena Expert | 1466 | 1467 |
| Humanity's Last Exam | 24.4% | — |
| SimpleQA Verified | 34.3% | — |
| Vectara Hallucination Rate | 14.2% | — |
Multimodal Too close to call
Kimi K2.5: 41.1 (#39), Qwen3.7 Plus: 41.8 (#33)
| Benchmark | Kimi K2.5 | Qwen3.7 Plus |
|---|---|---|
| LMArena Vision | 1269 | 1279 |
| LMArena Document | 1430 | 1444 |
Multilingual Too close to call
Kimi K2.5: 53.9 (#53), Qwen3.7 Plus: 54.8 (#38)
| Benchmark | Kimi K2.5 | Qwen3.7 Plus |
|---|---|---|
| LMArena Non-English | 1433 | 1445 |
| LMArena Chinese | 1495 | 1510 |
| LMArena French | 1454 | 1473 |
| LMArena German | 1441 | 1471 |
| LMArena Japanese | 1421 | 1413 |
| LMArena Korean | 1410 | 1415 |
| LMArena Russian | 1435 | 1457 |
| LMArena Spanish | 1450 | 1457 |
Instruction Following Too close to call
Kimi K2.5: 75.3 (#64), Qwen3.7 Plus: 75.8 (#52)
| Benchmark | Kimi K2.5 | Qwen3.7 Plus |
|---|---|---|
| LMArena Instruction Following | 1431 | 1440 |
Long Context Kimi K2.5 leads
Kimi K2.5: 52.1 (#7), Qwen3.7 Plus: 44.5 (#65)
| Benchmark | Kimi K2.5 | Qwen3.7 Plus |
|---|---|---|
| LMArena Longer Query | 1445 | 1455 |
| Fiction.LiveBench | 86.1% | — |
| CL-bench | 19.3% | — |
| CL-bench Life | 13.2% | — |
Writing & Preference Too close to call
Kimi K2.5: 65.1 (#53), Qwen3.7 Plus: 64.3 (#56)
| Benchmark | Kimi K2.5 | Qwen3.7 Plus |
|---|---|---|
| LMArena Text | 1445 | 1455 |
| LMArena Creative Writing | 1423 | 1439 |
| LMArena Multi-Turn | 1444 | 1460 |
| EQ-Bench Creative Writing | 1579 | — |
Frequently asked questions
Is Kimi K2.5 better than Qwen3.7 Plus?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 45.3 on the Noometry Index.
Which is cheaper, Kimi K2.5 or Qwen3.7 Plus?
Qwen3.7 Plus is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; Kimi K2.5 lists at $0.45 and $2.25.
Is Kimi K2.5 or Qwen3.7 Plus better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 36.6 in the Noometry coding category.
Which has the bigger context window?
Qwen3.7 Plus does, with 1M tokens against 262K.
How many benchmarks do Kimi K2.5 and Qwen3.7 Plus share?
26 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and Qwen3.7 Plus has 32.