Model comparison
Kimi K2.6 vs Qwen3 235B-A22B
Kimi K2.6 is the stronger model overall, scoring 47.5 to 43.2 on the Noometry Index.
Last verified . 33 shared benchmarks.
Alibaba (Qwen)Summary
- They share 33 benchmarks with published results for both. Kimi K2.6 scores higher in 7 categories and Qwen3 235B-A22B in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K2.6 leads 39.6 to 15.8.
- The biggest single-benchmark swing is WeirdML: 55.9% for Kimi K2.6 and 41% for Qwen3 235B-A22B.
- Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $0.95 / $4 for Kimi K2.6.
- Kimi K2.6 accepts more context: 262K tokens versus 131K.
Side by side
| Kimi K2.6 | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 47.5 | 43.2 |
| Released | 2026-04-20 | 2025-04 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 262K | 16K |
| Input $ / M tokens | $0.95 | $0.70 |
| Output $ / M tokens | $4 | $2.80 |
| Results tracked | 51 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
Kimi K2.6: 50.5 (#41), Qwen3 235B-A22B: 43.8 (#72)
| Benchmark | Kimi K2.6 | Qwen3 235B-A22B |
|---|---|---|
| SciCode | 53.5% | 42.4% |
| WeirdML | 55.9% | 41% |
| LMArena Coding | 1488 | 1445 |
| SWE-bench Verified | 76.7% | — |
| Aider Polyglot | — | 59.6% |
| LMArena WebDev | 1509 | — |
| ALE-Bench | 1,093 | — |
Agentic & Tool Use Qwen3 235B-A22B leads
Kimi K2.6: 21.8 (#128), Qwen3 235B-A22B: 33.7 (#48)
| Benchmark | Kimi K2.6 | Qwen3 235B-A22B |
|---|---|---|
| Vending-Bench 2 | 6,205 | -11.34 |
| Berkeley Function Calling Leaderboard | — | 52.1% |
| OSWorld 2.0 | 4.6% | — |
| ExploitBench | 18.4% | — |
| GBAEval | 0.9% | — |
| GDP.pdf | 12% | — |
Reasoning Kimi K2.6 leads
Kimi K2.6: 39.6 (#51), Qwen3 235B-A22B: 15.8 (#302)
| Benchmark | Kimi K2.6 | Qwen3 235B-A22B |
|---|---|---|
| CritPt | 8% | 0% |
| Chess Puzzles | 26% | 12% |
| LMArena Hard Prompts | 1470 | 1433 |
| Mystery Game Puzzles | 18% | 9% |
| DTBench | 90.9% | 80.3% |
| LMCA | 37.3% | 29.3% |
| Epoch Capabilities Index | 151.03 | 143.87 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 31% |
| Kagi LLM Benchmark | — | 69.4% |
| NYT Connections (extended) | 87.2% | — |
| ARC-AGI-1 | — | 11% |
| EBR-Bench | 2.4% | — |
| ForecastBench | — | 59.7 |
Math Kimi K2.6 leads
Kimi K2.6: 56.8 (#34), Qwen3 235B-A22B: 48.5 (#51)
| Benchmark | Kimi K2.6 | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 96.1% | 86.7% |
| LMArena Math | 1475 | 1431 |
| FrontierMath (Feb 2025 set) | 39% | 8.5% |
| FrontierMath Tier 4 (v1) | 14.6% | 0% |
| FrontierMath (Tiers 1-3) | 57.2% | — |
| FrontierMath Tier 4 | 25.6% | — |
| MathArena Final-Answer Competitions | 72.9% | — |
| ProofBench | 16% | — |
| Omni-MATH | — | 71.8% |
| MATH Level 5 | — | 68.9% |
Knowledge Kimi K2.6 leads
Kimi K2.6: 53.9 (#48), Qwen3 235B-A22B: 49.5 (#67)
| Benchmark | Kimi K2.6 | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 90.8% | 80.1% |
| SimpleQA Verified | 34.9% | 40.4% |
| Vectara Hallucination Rate | 10.8% | 9.3% |
| LMArena Expert | 1491 | 1460 |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| GPQA (HELM) | — | 72.7% |
Multimodal Not comparable
Kimi K2.6: 31.8 (#95), Qwen3 235B-A22B: —
| Benchmark | Kimi K2.6 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Vision | 1282 | — |
| Blueprint-Bench 2 | 3.9% | — |
| Furniture Assembly | 21.7% | — |
| LMArena Document | 1451 | — |
Multilingual Kimi K2.6 leads
Kimi K2.6: 55.1 (#35), Qwen3 235B-A22B: 52.5 (#84)
| Benchmark | Kimi K2.6 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1446 | 1410 |
| LMArena Chinese | 1521 | 1481 |
| LMArena French | 1471 | 1445 |
| LMArena German | 1449 | 1432 |
| LMArena Japanese | 1442 | 1400 |
| LMArena Korean | 1428 | 1391 |
| LMArena Russian | 1446 | 1411 |
| LMArena Spanish | 1463 | 1429 |
Instruction Following Kimi K2.6 leads
Kimi K2.6: 76.5 (#40), Qwen3 235B-A22B: 72.8 (#132)
| Benchmark | Kimi K2.6 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1451 | 1408 |
| IFEval | — | 83.5% |
Long Context Qwen3 235B-A22B leads
Kimi K2.6: 44.9 (#50), Qwen3 235B-A22B: 46.1 (#25)
| Benchmark | Kimi K2.6 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1468 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Kimi K2.6 leads
Kimi K2.6: 68.7 (#25), Qwen3 235B-A22B: 59.8 (#103)
| Benchmark | Kimi K2.6 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1455 | 1419 |
| LMArena Creative Writing | 1434 | 1384 |
| EQ-Bench Creative Writing | 1725 | 1366 |
| LMArena Multi-Turn | 1453 | 1432 |
| Short-Story Creative Writing | — | 83% |
| WildBench | — | 86.6% |
| EQ-Bench 4 | 1202 | — |
Frequently asked questions
Is Kimi K2.6 better than Qwen3 235B-A22B?
Kimi K2.6 is the stronger model overall, scoring 47.5 to 43.2 on the Noometry Index.
Which is cheaper, Kimi K2.6 or Qwen3 235B-A22B?
Qwen3 235B-A22B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; Kimi K2.6 lists at $0.95 and $4.
Is Kimi K2.6 or Qwen3 235B-A22B better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.5 versus 43.8 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.6 does, with 262K tokens against 131K.
How many benchmarks do Kimi K2.6 and Qwen3 235B-A22B share?
33 benchmarks have published results for both models. Kimi K2.6 has 51 scored results on Noometry and Qwen3 235B-A22B has 49.