Model comparison
Kimi K2 (Jul 2025) vs Qwen3.5 397B-A17B
Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 40.9 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Kimi K2 (Jul 2025) scores higher in 1 category and Qwen3.5 397B-A17B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5 397B-A17B leads 53.3 to 36.9.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 64.4% for Kimi K2 (Jul 2025) and 73.7% for Qwen3.5 397B-A17B.
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $0.60 / $3.60 for Qwen3.5 397B-A17B.
Side by side
| Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B | |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 40.9 | 45.7 |
| Released | 2025-07-12 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 262K | 66K |
| Input $ / M tokens | $0.57 | $0.60 |
| Output $ / M tokens | $2.30 | $3.60 |
| Results tracked | 42 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 43.1 (#83), Qwen3.5 397B-A17B: 41.7 (#108)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Coding | 1399 | 1464 |
| SWE-bench Verified (bash only) | 63.4% | — |
| Aider Polyglot | 59.1% | — |
| LMArena WebDev | — | 1400 |
| GSO | 4.9% | — |
| WeirdML | 42.8% | — |
| ALE-Bench | 597.5 | — |
Agentic & Tool Use Qwen3.5 397B-A17B leads
Kimi K2 (Jul 2025): 32.2 (#59), Qwen3.5 397B-A17B: 33.2 (#51)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B |
|---|---|---|
| Terminal-Bench | 35.7% | — |
| APEX-Agents | — | 24.9% |
| Berkeley Function Calling Leaderboard | 59.1% | — |
| τ²-bench Airline | — | 81.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 84.4% |
| τ²-bench Telecom | — | 97.8% |
| METR Time Horizons | 59.2% | — |
Reasoning Qwen3.5 397B-A17B leads
Kimi K2 (Jul 2025): 22.7 (#173), Qwen3.5 397B-A17B: 33.6 (#62)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B |
|---|---|---|
| Kagi LLM Benchmark | 64.4% | 73.7% |
| LMArena Hard Prompts | 1384 | 1448 |
| Epoch Capabilities Index | 146.02 | 146.64 |
| SimpleBench | 26.3% | — |
| NYT Connections (extended) | — | 58.9% |
| Chess Puzzles | — | 13% |
| Thematic Generalization | — | 65.1% |
| Mystery Game Puzzles | — | 18% |
| DTBench | — | 87.5% |
| LMCA | — | 37.9% |
| ForecastBench | 60.2 | — |
Math Qwen3.5 397B-A17B leads
Kimi K2 (Jul 2025): 40.1 (#81), Qwen3.5 397B-A17B: 44.9 (#64)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Math | 1396 | 1455 |
| FrontierMath (Tiers 1-3) | — | 31.2% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| Omni-MATH | 65.4% | — |
| FrontierMath (Feb 2025 set) | 21.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Qwen3.5 397B-A17B leads
Kimi K2 (Jul 2025): 36.9 (#153), Qwen3.5 397B-A17B: 53.3 (#53)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Expert | 1363 | 1461 |
| GPQA Diamond | — | 86.4% |
| MMLU-Pro | 81.9% | — |
| Confabulations | 20.4% | — |
| Vectara Hallucination Rate | 17.9% | — |
| GPQA (HELM) | 65.3% | — |
Multimodal Not comparable
Kimi K2 (Jul 2025): —, Qwen3.5 397B-A17B: 40.0 (#42)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | — | 1262 |
Multilingual Qwen3.5 397B-A17B leads
Kimi K2 (Jul 2025): 49.8 (#125), Qwen3.5 397B-A17B: 54.0 (#56)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | 1372 | 1431 |
| LMArena Chinese | 1416 | 1501 |
| LMArena French | 1379 | 1461 |
| LMArena German | 1386 | 1447 |
| LMArena Japanese | 1349 | 1427 |
| LMArena Korean | 1324 | 1382 |
| LMArena Russian | 1384 | 1429 |
| LMArena Spanish | 1383 | 1443 |
Instruction Following Qwen3.5 397B-A17B leads
Kimi K2 (Jul 2025): 71.3 (#151), Qwen3.5 397B-A17B: 75.2 (#73)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | 1347 | 1424 |
| IFEval | 85% | — |
Long Context Qwen3.5 397B-A17B leads
Kimi K2 (Jul 2025): 41.2 (#140), Qwen3.5 397B-A17B: 44.1 (#71)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | 1353 | 1442 |
| Fiction.LiveBench | 66.7% | — |
| CL-bench | 17.6% | — |
Writing & Preference Too close to call
Kimi K2 (Jul 2025): 62.5 (#76), Qwen3.5 397B-A17B: 62.5 (#74)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | 1380 | 1438 |
| LMArena Creative Writing | 1349 | 1401 |
| EQ-Bench Creative Writing | 1666 | 1478 |
| LMArena Multi-Turn | 1371 | 1447 |
| Short-Story Creative Writing | 85.6% | — |
| WildBench | 86.2% | — |
Frequently asked questions
Is Kimi K2 (Jul 2025) better than Qwen3.5 397B-A17B?
Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 40.9 on the Noometry Index.
Which is cheaper, Kimi K2 (Jul 2025) or Qwen3.5 397B-A17B?
Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; Qwen3.5 397B-A17B lists at $0.60 and $3.60.
Is Kimi K2 (Jul 2025) or Qwen3.5 397B-A17B better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 43.1 versus 41.7 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Kimi K2 (Jul 2025) and Qwen3.5 397B-A17B share?
20 benchmarks have published results for both models. Kimi K2 (Jul 2025) has 42 scored results on Noometry and Qwen3.5 397B-A17B has 36.