Model comparison
Kimi K3 vs Qwen3 Max
Kimi K3 is the stronger model overall, scoring 59.5 to 43.7 on the Noometry Index. Qwen3 Max costs 2.5× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
Alibaba (Qwen)Summary
- They share 29 benchmarks with published results for both. Kimi K3 scores higher in 8 categories and Qwen3 Max in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 22.6.
- The biggest single-benchmark swing is NYT Connections (extended): 93.6% for Kimi K3 and 30.1% for Qwen3 Max.
- Qwen3 Max is cheaper at $1.20 / $6 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 262K.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| Kimi K3 | Qwen3 Max | |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 59.5 | 43.7 |
| Released | 2026-07-16 | 2025-09-23 |
| Weights | Open | Proprietary |
| Context window | 1.05M | 262K |
| Max output | 1.05M | 66K |
| Input $ / M tokens | $3 | $1.20 |
| Output $ / M tokens | $15 | $6 |
| Results tracked | 53 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Kimi K3: 61.0 (#10), Qwen3 Max: 43.0 (#93)
| Benchmark | Kimi K3 | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1508 | 1456 |
| ALE-Bench | 1,524 | 370.45 |
| DeepSWE | 68.5% | — |
| FrontierCode | 44.2% | — |
| LMArena WebDev | 1654 | — |
| FrontierSWE | 25.9% | — |
| SciCode | 59.5% | — |
| WeirdML | 82.6% | — |
Agentic & Tool Use Not comparable
Kimi K3: 41.8 (#20), Qwen3 Max: —
| Benchmark | Kimi K3 | Qwen3 Max |
|---|---|---|
| Vending-Bench 2 | 5,165 | 71.56 |
| APEX-Agents | 50.6% | — |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 32% | — |
| GBAEval | 48.3% | — |
| GDP.pdf | 19% | — |
Reasoning Kimi K3 leads
Kimi K3: 63.0 (#17), Qwen3 Max: 22.6 (#190)
| Benchmark | Kimi K3 | Qwen3 Max |
|---|---|---|
| NYT Connections (extended) | 93.6% | 30.1% |
| Chess Puzzles | 39% | 4% |
| LMArena Hard Prompts | 1496 | 1448 |
| Mystery Game Puzzles | 26% | 5% |
| DTBench | 91.2% | 82.1% |
| LMCA | 52.7% | 28.3% |
| Epoch Capabilities Index | 157.45 | 142.38 |
| ARC-AGI-2 | 60.4% | — |
| SimpleBench | 60.7% | — |
| Kagi LLM Benchmark | — | 72.5% |
| ARC-AGI-1 | 94.5% | — |
| CritPt | 23.4% | — |
| Surface Evolver Bench | 95% | — |
| ForecastBench | 61.1 | — |
Math Kimi K3 leads
Kimi K3: 74.2 (#16), Qwen3 Max: 38.7 (#131)
| Benchmark | Kimi K3 | Qwen3 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 72.2% | 18.9% |
| OTIS Mock AIME 2024-2025 | 97.2% | 73.3% |
| LMArena Math | 1491 | 1446 |
| FrontierMath Tier 4 | 39% | — |
| MathArena Final-Answer Competitions | 87.8% | — |
| ProofBench | 87% | — |
| MATH Level 5 | — | 97.1% |
Knowledge Kimi K3 leads
Kimi K3: 63.2 (#21), Qwen3 Max: 48.1 (#78)
| Benchmark | Kimi K3 | Qwen3 Max |
|---|---|---|
| GPQA Diamond | 93.1% | 72.6% |
| SimpleQA Verified | 50.6% | 48.7% |
| LMArena Expert | 1521 | 1455 |
Multimodal Not comparable
Kimi K3: 37.8 (#70), Qwen3 Max: —
| Benchmark | Kimi K3 | Qwen3 Max |
|---|---|---|
| Blueprint-Bench 2 | 29.5% | — |
| Furniture Assembly | 34.2% | — |
Multilingual Kimi K3 leads
Kimi K3: 56.3 (#21), Qwen3 Max: 53.7 (#62)
| Benchmark | Kimi K3 | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1466 | 1429 |
| LMArena Chinese | 1529 | 1478 |
| LMArena French | 1491 | 1449 |
| LMArena German | 1488 | 1463 |
| LMArena Japanese | 1487 | 1397 |
| LMArena Korean | 1458 | 1399 |
| LMArena Russian | 1482 | 1428 |
| LMArena Spanish | 1472 | 1462 |
Instruction Following Kimi K3 leads
Kimi K3: 77.7 (#14), Qwen3 Max: 74.8 (#87)
| Benchmark | Kimi K3 | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1483 | 1419 |
Long Context Kimi K3 leads
Kimi K3: 45.8 (#29), Qwen3 Max: 41.6 (#134)
| Benchmark | Kimi K3 | Qwen3 Max |
|---|---|---|
| LMArena Longer Query | 1494 | 1438 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 14.5% |
Writing & Preference Kimi K3 leads
Kimi K3: 76.6 (#4), Qwen3 Max: 62.4 (#76)
| Benchmark | Kimi K3 | Qwen3 Max |
|---|---|---|
| LMArena Text | 1476 | 1439 |
| LMArena Creative Writing | 1454 | 1402 |
| LMArena Multi-Turn | 1488 | 1446 |
| EQ-Bench Creative Writing | 2082 | — |
| EQ-Bench 4 | 1339 | — |
Frequently asked questions
Is Kimi K3 better than Qwen3 Max?
Kimi K3 is the stronger model overall, scoring 59.5 to 43.7 on the Noometry Index. Qwen3 Max costs 2.5× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, Kimi K3 or Qwen3 Max?
Qwen3 Max is cheaper. It lists at $1.20 per million input tokens and $6 per million output tokens; Kimi K3 lists at $3 and $15.
Is Kimi K3 or Qwen3 Max better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 43.0 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 262K.
How many benchmarks do Kimi K3 and Qwen3 Max share?
29 benchmarks have published results for both models. Kimi K3 has 53 scored results on Noometry and Qwen3 Max has 33.