Model comparison
Kimi K2 (Jul 2025) vs Qwen3 32B
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 39.2 on the Noometry Index.
Last verified . 19 shared benchmarks.
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. Kimi K2 (Jul 2025) scores higher in 6 categories and Qwen3 32B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Kimi K2 (Jul 2025) leads 62.3 to 52.9.
- The biggest single-benchmark swing is Aider Polyglot: 59.1% for Kimi K2 (Jul 2025) and 40% for Qwen3 32B.
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $0.70 / $2.80 for Qwen3 32B.
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 131K.
Side by side
| Kimi K2 (Jul 2025) | Qwen3 32B | |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 41.2 | 39.2 |
| Released | 2025-07-12 | 2025-04 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 262K | 16K |
| Input $ / M tokens | $0.57 | $0.70 |
| Output $ / M tokens | $2.30 | $2.80 |
| Results tracked | 42 | 26 |
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): 42.4 (#102), Qwen3 32B: 37.7 (#190)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3 32B |
|---|---|---|
| Aider Polyglot | 59.1% | 40% |
| LMArena Coding | 1399 | 1358 |
| SWE-bench Verified (bash only) | 63.4% | — |
| SciCode | — | 35.4% |
| GSO | 4.9% | — |
| WeirdML | 42.8% | — |
| ALE-Bench | 597.5 | — |
Agentic & Tool Use Too close to call
Kimi K2 (Jul 2025): 32.4 (#64), Qwen3 32B: 32.6 (#62)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3 32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 59.1% | 48.7% |
| Terminal-Bench | 35.7% | — |
| METR Time Horizons | 59.2% | — |
Reasoning Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 23.3 (#179), Qwen3 32B: 20.2 (#241)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3 32B |
|---|---|---|
| Kagi LLM Benchmark | 64.4% | 54.9% |
| LMArena Hard Prompts | 1384 | 1334 |
| Epoch Capabilities Index | 146.01 | 138.51 |
| SimpleBench | 26.3% | — |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 5% |
| DTBench | — | 67.5% |
| LMCA | — | 17.3% |
| ForecastBench | 60.2 | — |
Math Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 42.7 (#83), Qwen3 32B: 39.7 (#99)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3 32B |
|---|---|---|
| LMArena Math | 1397 | 1399 |
| OTIS Mock AIME 2024-2025 | — | 66.9% |
| Omni-MATH | 65.4% | — |
| FrontierMath (Feb 2025 set) | 21.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Qwen3 32B leads
Kimi K2 (Jul 2025): 37.3 (#157), Qwen3 32B: 40.0 (#125)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3 32B |
|---|---|---|
| Vectara Hallucination Rate | 17.9% | 5.9% |
| LMArena Expert | 1365 | 1362 |
| GPQA Diamond | — | 65.7% |
| MMLU-Pro | 81.9% | — |
| Confabulations | 20.4% | — |
| GPQA (HELM) | 65.3% | — |
Multilingual Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 49.6 (#130), Qwen3 32B: 45.6 (#167)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3 32B |
|---|---|---|
| LMArena Non-English | 1372 | 1317 |
| LMArena Chinese | 1415 | 1357 |
| LMArena German | 1387 | 1341 |
| LMArena Russian | 1385 | 1311 |
| LMArena French | 1379 | — |
| LMArena Japanese | 1349 | — |
| LMArena Korean | 1325 | — |
| LMArena Spanish | 1386 | — |
Instruction Following Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 71.1 (#156), Qwen3 32B: 68.9 (#179)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | 1348 | 1305 |
| IFEval | 85% | — |
Long Context Qwen3 32B leads
Kimi K2 (Jul 2025): 41.2 (#145), Qwen3 32B: 43.8 (#87)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3 32B |
|---|---|---|
| Fiction.LiveBench | 66.7% | 74.2% |
| LMArena Longer Query | 1353 | 1327 |
| CL-bench | 17.6% | — |
Writing & Preference Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 62.3 (#78), Qwen3 32B: 52.9 (#163)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3 32B |
|---|---|---|
| LMArena Text | 1380 | 1340 |
| LMArena Creative Writing | 1350 | 1297 |
| LMArena Multi-Turn | 1371 | 1331 |
| Short-Story Creative Writing | 85.6% | — |
| EQ-Bench Creative Writing | 1666 | — |
| WildBench | 86.2% | — |
Frequently asked questions
Is Kimi K2 (Jul 2025) better than Qwen3 32B?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 39.2 on the Noometry Index.
Which is cheaper, Kimi K2 (Jul 2025) or Qwen3 32B?
Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; Qwen3 32B lists at $0.70 and $2.80.
Is Kimi K2 (Jul 2025) or Qwen3 32B better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 37.7 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 131K.
How many benchmarks do Kimi K2 (Jul 2025) and Qwen3 32B share?
19 benchmarks have published results for both models. Kimi K2 (Jul 2025) has 42 scored results on Noometry and Qwen3 32B has 26.