Model comparison
Kimi K2.5 vs Qwen3.5 397B-A17B
Kimi K2.5 is the stronger model overall, scoring 48.1 to 46.0 on the Noometry Index.
Last verified . 27 shared benchmarks.
Alibaba (Qwen)Summary
- They share 27 benchmarks with published results for both. Kimi K2.5 scores higher in 9 categories and Qwen3.5 397B-A17B in 1 category; 5 gaps are clear of the uncertainty.
- The widest gap is in long context, where Kimi K2.5 leads 52.1 to 44.1.
- The biggest single-benchmark swing is NYT Connections (extended): 69.9% for Kimi K2.5 and 58.9% for Qwen3.5 397B-A17B.
- Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $0.60 / $3.60 for Qwen3.5 397B-A17B.
Side by side
| Kimi K2.5 | Qwen3.5 397B-A17B | |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 48.1 | 46.0 |
| Released | 2026-01-27 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 262K | 66K |
| Input $ / M tokens | $0.45 | $0.60 |
| Output $ / M tokens | $2.25 | $3.60 |
| Results tracked | 51 | 36 |
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), Qwen3.5 397B-A17B: 42.0 (#114)
| Benchmark | Kimi K2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena WebDev | 1437 | 1400 |
| LMArena Coding | 1474 | 1465 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 70.8% | — |
| SWE-bench Multilingual | 67.3% | — |
| SciCode | 49% | — |
| WeirdML | 45.6% | — |
| ALE-Bench | 821.65 | — |
Agentic & Tool Use Too close to call
Kimi K2.5: 34.2 (#48), Qwen3.5 397B-A17B: 33.3 (#53)
| Benchmark | Kimi K2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| Terminal-Bench | 43.2% | — |
| APEX-Agents | — | 24.9% |
| τ²-bench Airline | — | 81.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 84.4% |
| τ²-bench Telecom | — | 97.8% |
| OSWorld | 63.3% | — |
| Vending-Bench 2 | 1,198 | — |
Reasoning Qwen3.5 397B-A17B leads
Kimi K2.5: 31.2 (#80), Qwen3.5 397B-A17B: 34.5 (#70)
| Benchmark | Kimi K2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| Kagi LLM Benchmark | 78.5% | 73.7% |
| NYT Connections (extended) | 69.9% | 58.9% |
| Chess Puzzles | 12% | 13% |
| Thematic Generalization | 69.4% | 65.1% |
| LMArena Hard Prompts | 1453 | 1448 |
| Epoch Capabilities Index | 148.03 | 146.65 |
| ARC-AGI-2 | 11.8% | — |
| SimpleBench | 46.8% | — |
| ARC-AGI-1 | 65.3% | — |
| CritPt | 3.1% | — |
| EnigmaEval | 3.4% | — |
| Mystery Game Puzzles | — | 18% |
| DTBench | — | 87.5% |
| LMCA | — | 37.9% |
Math Kimi K2.5 leads
Kimi K2.5: 51.8 (#53), Qwen3.5 397B-A17B: 46.1 (#73)
| Benchmark | Kimi K2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 88.9% |
| LMArena Math | 1470 | 1454 |
| FrontierMath (Tiers 1-3) | — | 31.2% |
| MathArena Final-Answer Competitions | 62.3% | — |
| FrontierMath (Feb 2025 set) | 27.9% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Too close to call
Kimi K2.5: 53.6 (#56), Qwen3.5 397B-A17B: 53.3 (#58)
| Benchmark | Kimi K2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| GPQA Diamond | 87.6% | 86.4% |
| LMArena Expert | 1466 | 1462 |
| 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), Qwen3.5 397B-A17B: 40.7 (#44)
| Benchmark | Kimi K2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | 1269 | 1263 |
| LMArena Document | 1430 | — |
Multilingual Too close to call
Kimi K2.5: 53.9 (#53), Qwen3.5 397B-A17B: 53.7 (#59)
| Benchmark | Kimi K2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | 1433 | 1430 |
| LMArena Chinese | 1495 | 1500 |
| LMArena French | 1454 | 1461 |
| LMArena German | 1441 | 1447 |
| LMArena Japanese | 1421 | 1426 |
| LMArena Korean | 1410 | 1384 |
| LMArena Russian | 1435 | 1429 |
| LMArena Spanish | 1450 | 1441 |
Instruction Following Too close to call
Kimi K2.5: 75.3 (#64), Qwen3.5 397B-A17B: 75.0 (#77)
| Benchmark | Kimi K2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | 1431 | 1424 |
Long Context Kimi K2.5 leads
Kimi K2.5: 52.1 (#7), Qwen3.5 397B-A17B: 44.1 (#74)
| Benchmark | Kimi K2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | 1445 | 1442 |
| Fiction.LiveBench | 86.1% | — |
| CL-bench | 19.3% | — |
| CL-bench Life | 13.2% | — |
Writing & Preference Kimi K2.5 leads
Kimi K2.5: 65.1 (#53), Qwen3.5 397B-A17B: 62.3 (#79)
| Benchmark | Kimi K2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | 1445 | 1438 |
| LMArena Creative Writing | 1423 | 1401 |
| EQ-Bench Creative Writing | 1579 | 1478 |
| LMArena Multi-Turn | 1444 | 1446 |
Frequently asked questions
Is Kimi K2.5 better than Qwen3.5 397B-A17B?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 46.0 on the Noometry Index.
Which is cheaper, Kimi K2.5 or Qwen3.5 397B-A17B?
Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; Qwen3.5 397B-A17B lists at $0.60 and $3.60.
Is Kimi K2.5 or Qwen3.5 397B-A17B better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 42.0 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Kimi K2.5 and Qwen3.5 397B-A17B share?
27 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and Qwen3.5 397B-A17B has 36.