Model comparison
Kimi K2.5 vs Qwen3.6 Max Preview
Qwen3.6 Max Preview is the stronger model overall, scoring 51.5 to 48.1 on the Noometry Index. Kimi K2.5 costs 3.2× less per token, which makes it the better buy when Qwen3.6 Max Preview's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. Kimi K2.5 scores higher in 3 categories and Qwen3.6 Max Preview in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.6 Max Preview leads 41.7 to 31.2.
- The biggest single-benchmark swing is SimpleQA Verified: 34.3% for Kimi K2.5 and 52% for Qwen3.6 Max Preview.
- Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $1.30 / $7.80 for Qwen3.6 Max Preview.
- Kimi K2.5 has downloadable open weights; the other is API-only.
Side by side
| Kimi K2.5 | Qwen3.6 Max Preview | |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 48.1 | 51.5 |
| Released | 2026-01-27 | 2026-04-20 |
| Weights | Open | Proprietary |
| Context window | 262K | 262K |
| Max output | 262K | 66K |
| Input $ / M tokens | $0.45 | $1.30 |
| Output $ / M tokens | $2.25 | $7.80 |
| Results tracked | 51 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Kimi K2.5: 48.8 (#53), Qwen3.6 Max Preview: 48.7 (#54)
| Benchmark | Kimi K2.5 | Qwen3.6 Max Preview |
|---|---|---|
| SWE-bench Verified | 73.8% | 76.7% |
| LMArena WebDev | 1437 | 1482 |
| LMArena Coding | 1474 | 1471 |
| SWE-bench Verified (bash only) | 70.8% | — |
| SWE-bench Multilingual | 67.3% | — |
| SciCode | 49% | — |
| WeirdML | 45.6% | — |
| ALE-Bench | 821.65 | — |
Agentic & Tool Use Not comparable
Kimi K2.5: 34.2 (#48), Qwen3.6 Max Preview: —
| Benchmark | Kimi K2.5 | Qwen3.6 Max Preview |
|---|---|---|
| Vending-Bench 2 | 1,198 | 4,254 |
| Terminal-Bench | 43.2% | — |
| OSWorld | 63.3% | — |
Reasoning Qwen3.6 Max Preview leads
Kimi K2.5: 31.2 (#80), Qwen3.6 Max Preview: 41.7 (#53)
| Benchmark | Kimi K2.5 | Qwen3.6 Max Preview |
|---|---|---|
| SimpleBench | 46.8% | 63% |
| NYT Connections (extended) | 69.9% | 74.1% |
| Chess Puzzles | 12% | 20% |
| LMArena Hard Prompts | 1453 | 1457 |
| Epoch Capabilities Index | 148.03 | 149.24 |
| ARC-AGI-2 | 11.8% | — |
| Kagi LLM Benchmark | 78.5% | — |
| ARC-AGI-1 | 65.3% | — |
| CritPt | 3.1% | — |
| EnigmaEval | 3.4% | — |
| Thematic Generalization | 69.4% | — |
| Mystery Game Puzzles | — | 19% |
| DTBench | — | 87.2% |
| LMCA | — | 42.5% |
Math Qwen3.6 Max Preview leads
Kimi K2.5: 51.8 (#53), Qwen3.6 Max Preview: 54.1 (#46)
| Benchmark | Kimi K2.5 | Qwen3.6 Max Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 91.1% |
| LMArena Math | 1470 | 1465 |
| FrontierMath (Feb 2025 set) | 27.9% | 23.1% |
| FrontierMath Tier 4 (v1) | 4.2% | 4.2% |
| MathArena Final-Answer Competitions | 62.3% | — |
Knowledge Qwen3.6 Max Preview leads
Kimi K2.5: 53.6 (#56), Qwen3.6 Max Preview: 57.6 (#39)
| Benchmark | Kimi K2.5 | Qwen3.6 Max Preview |
|---|---|---|
| GPQA Diamond | 87.6% | 87.4% |
| SimpleQA Verified | 34.3% | 52% |
| LMArena Expert | 1466 | 1478 |
| Humanity's Last Exam | 24.4% | — |
| Vectara Hallucination Rate | 14.2% | — |
Multimodal Not comparable
Kimi K2.5: 41.1 (#39), Qwen3.6 Max Preview: —
| Benchmark | Kimi K2.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Vision | 1269 | — |
| LMArena Document | 1430 | — |
Multilingual Too close to call
Kimi K2.5: 53.9 (#53), Qwen3.6 Max Preview: 54.2 (#48)
| Benchmark | Kimi K2.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Non-English | 1433 | 1437 |
| LMArena Chinese | 1495 | 1487 |
| LMArena French | 1454 | 1449 |
| LMArena Russian | 1435 | 1445 |
| LMArena Spanish | 1450 | 1454 |
| LMArena German | 1441 | — |
| LMArena Japanese | 1421 | — |
| LMArena Korean | 1410 | — |
Instruction Following Too close to call
Kimi K2.5: 75.3 (#64), Qwen3.6 Max Preview: 75.7 (#55)
| Benchmark | Kimi K2.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Instruction Following | 1431 | 1438 |
Long Context Kimi K2.5 leads
Kimi K2.5: 52.1 (#7), Qwen3.6 Max Preview: 44.6 (#61)
| Benchmark | Kimi K2.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Longer Query | 1445 | 1457 |
| 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), Qwen3.6 Max Preview: 63.8 (#60)
| Benchmark | Kimi K2.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Text | 1445 | 1447 |
| LMArena Creative Writing | 1423 | 1435 |
| LMArena Multi-Turn | 1444 | 1456 |
| EQ-Bench Creative Writing | 1579 | — |
Frequently asked questions
Is Kimi K2.5 better than Qwen3.6 Max Preview?
Qwen3.6 Max Preview is the stronger model overall, scoring 51.5 to 48.1 on the Noometry Index. Kimi K2.5 costs 3.2× less per token, which makes it the better buy when Qwen3.6 Max Preview's lead doesn't matter for your workload.
Which is cheaper, Kimi K2.5 or Qwen3.6 Max Preview?
Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; Qwen3.6 Max Preview lists at $1.30 and $7.80.
Is Kimi K2.5 or Qwen3.6 Max Preview better for coding?
They score almost the same on coding (48.8 vs 48.7); test both on your own repository before choosing.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Kimi K2.5 and Qwen3.6 Max Preview share?
26 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and Qwen3.6 Max Preview has 29.