Model comparison
Kimi K2.5 vs Qwen3 32B
Kimi K2.5 is the stronger model overall, scoring 48.1 to 39.2 on the Noometry Index.
Last verified . 22 shared benchmarks.
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. Kimi K2.5 scores higher in 9 categories and Qwen3 32B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Kimi K2.5 leads 53.6 to 40.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 92.2% for Kimi K2.5 and 66.9% for Qwen3 32B.
- Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $0.70 / $2.80 for Qwen3 32B.
- Kimi K2.5 accepts more context: 262K tokens versus 131K.
Side by side
| Kimi K2.5 | Qwen3 32B | |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 48.1 | 39.2 |
| Released | 2026-01-27 | 2025-04 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 262K | 16K |
| Input $ / M tokens | $0.45 | $0.70 |
| Output $ / M tokens | $2.25 | $2.80 |
| Results tracked | 51 | 26 |
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 32B: 37.7 (#190)
| Benchmark | Kimi K2.5 | Qwen3 32B |
|---|---|---|
| SciCode | 49% | 35.4% |
| LMArena Coding | 1474 | 1358 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 70.8% | — |
| Aider Polyglot | — | 40% |
| 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 32B: 32.6 (#62)
| Benchmark | Kimi K2.5 | Qwen3 32B |
|---|---|---|
| Terminal-Bench | 43.2% | — |
| Berkeley Function Calling Leaderboard | — | 48.7% |
| OSWorld | 63.3% | — |
| Vending-Bench 2 | 1,198 | — |
Reasoning Kimi K2.5 leads
Kimi K2.5: 31.2 (#80), Qwen3 32B: 20.2 (#241)
| Benchmark | Kimi K2.5 | Qwen3 32B |
|---|---|---|
| Kagi LLM Benchmark | 78.5% | 54.9% |
| CritPt | 3.1% | 0.3% |
| Chess Puzzles | 12% | 5% |
| LMArena Hard Prompts | 1453 | 1334 |
| Epoch Capabilities Index | 148.03 | 138.51 |
| ARC-AGI-2 | 11.8% | — |
| SimpleBench | 46.8% | — |
| NYT Connections (extended) | 69.9% | — |
| ARC-AGI-1 | 65.3% | — |
| EnigmaEval | 3.4% | — |
| Thematic Generalization | 69.4% | — |
| DTBench | — | 67.5% |
| LMCA | — | 17.3% |
Math Kimi K2.5 leads
Kimi K2.5: 51.8 (#53), Qwen3 32B: 39.7 (#99)
| Benchmark | Kimi K2.5 | Qwen3 32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 66.9% |
| LMArena Math | 1470 | 1399 |
| MathArena Final-Answer Competitions | 62.3% | — |
| FrontierMath (Feb 2025 set) | 27.9% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Kimi K2.5 leads
Kimi K2.5: 53.6 (#56), Qwen3 32B: 40.0 (#125)
| Benchmark | Kimi K2.5 | Qwen3 32B |
|---|---|---|
| GPQA Diamond | 87.6% | 65.7% |
| Vectara Hallucination Rate | 14.2% | 5.9% |
| LMArena Expert | 1466 | 1362 |
| Humanity's Last Exam | 24.4% | — |
| SimpleQA Verified | 34.3% | — |
Multimodal Not comparable
Kimi K2.5: 41.1 (#39), Qwen3 32B: —
| Benchmark | Kimi K2.5 | Qwen3 32B |
|---|---|---|
| LMArena Vision | 1269 | — |
| LMArena Document | 1430 | — |
Multilingual Kimi K2.5 leads
Kimi K2.5: 53.9 (#53), Qwen3 32B: 45.6 (#167)
| Benchmark | Kimi K2.5 | Qwen3 32B |
|---|---|---|
| LMArena Non-English | 1433 | 1317 |
| LMArena Chinese | 1495 | 1357 |
| LMArena German | 1441 | 1341 |
| LMArena Russian | 1435 | 1311 |
| LMArena French | 1454 | — |
| LMArena Japanese | 1421 | — |
| LMArena Korean | 1410 | — |
| LMArena Spanish | 1450 | — |
Instruction Following Kimi K2.5 leads
Kimi K2.5: 75.3 (#64), Qwen3 32B: 68.9 (#179)
| Benchmark | Kimi K2.5 | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | 1431 | 1305 |
Long Context Kimi K2.5 leads
Kimi K2.5: 52.1 (#7), Qwen3 32B: 43.8 (#87)
| Benchmark | Kimi K2.5 | Qwen3 32B |
|---|---|---|
| Fiction.LiveBench | 86.1% | 74.2% |
| LMArena Longer Query | 1445 | 1327 |
| CL-bench | 19.3% | — |
| CL-bench Life | 13.2% | — |
Writing & Preference Kimi K2.5 leads
Kimi K2.5: 65.1 (#53), Qwen3 32B: 52.9 (#163)
| Benchmark | Kimi K2.5 | Qwen3 32B |
|---|---|---|
| LMArena Text | 1445 | 1340 |
| LMArena Creative Writing | 1423 | 1297 |
| LMArena Multi-Turn | 1444 | 1331 |
| EQ-Bench Creative Writing | 1579 | — |
Frequently asked questions
Is Kimi K2.5 better than Qwen3 32B?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 39.2 on the Noometry Index.
Which is cheaper, Kimi K2.5 or Qwen3 32B?
Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; Qwen3 32B lists at $0.70 and $2.80.
Is Kimi K2.5 or Qwen3 32B better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 37.7 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.5 does, with 262K tokens against 131K.
How many benchmarks do Kimi K2.5 and Qwen3 32B share?
22 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and Qwen3 32B has 26.