Model comparison
Kimi K2 (Jul 2025) vs Qwen3-Next 80B-A3B Instruct
Qwen3-Next 80B-A3B Instruct is the stronger model overall, scoring 43.0 to 41.2 on the Noometry Index.
Last verified . 25 shared benchmarks.
Alibaba (Qwen)Summary
- They share 25 benchmarks with published results for both. Kimi K2 (Jul 2025) scores higher in 4 categories and Qwen3-Next 80B-A3B Instruct in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3-Next 80B-A3B Instruct leads 31.1 to 23.3.
- The biggest single-benchmark swing is Omni-MATH: 65.4% for Kimi K2 (Jul 2025) and 46.7% for Qwen3-Next 80B-A3B Instruct.
- Qwen3-Next 80B-A3B Instruct is cheaper at $0.50 / $2 per million input/output tokens, against $0.57 / $2.30 for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 131K.
Side by side
| Kimi K2 (Jul 2025) | Qwen3-Next 80B-A3B Instruct | |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 41.2 | 43.0 |
| Released | 2025-07-12 | 2025-09 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 262K | 33K |
| Input $ / M tokens | $0.57 | $0.50 |
| Output $ / M tokens | $2.30 | $2 |
| Results tracked | 42 | 25 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Kimi K2 (Jul 2025): 42.4 (#102), Qwen3-Next 80B-A3B Instruct: 42.5 (#98)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Coding | 1399 | 1440 |
| SWE-bench Verified (bash only) | 63.4% | — |
| Aider Polyglot | 59.1% | — |
| GSO | 4.9% | — |
| WeirdML | 42.8% | — |
| ALE-Bench | 597.5 | — |
Agentic & Tool Use Not comparable
Kimi K2 (Jul 2025): 32.4 (#64), Qwen3-Next 80B-A3B Instruct: —
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| Terminal-Bench | 35.7% | — |
| Berkeley Function Calling Leaderboard | 59.1% | — |
| METR Time Horizons | 59.2% | — |
Reasoning Qwen3-Next 80B-A3B Instruct leads
Kimi K2 (Jul 2025): 23.3 (#179), Qwen3-Next 80B-A3B Instruct: 31.1 (#81)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| Kagi LLM Benchmark | 64.4% | 66.7% |
| LMArena Hard Prompts | 1384 | 1428 |
| SimpleBench | 26.3% | — |
| Epoch Capabilities Index | 146.01 | — |
| ForecastBench | 60.2 | — |
Math Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 42.7 (#83), Qwen3-Next 80B-A3B Instruct: 38.8 (#126)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| Omni-MATH | 65.4% | 46.7% |
| LMArena Math | 1397 | 1440 |
| FrontierMath (Feb 2025 set) | 21.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Qwen3-Next 80B-A3B Instruct leads
Kimi K2 (Jul 2025): 37.3 (#157), Qwen3-Next 80B-A3B Instruct: 41.8 (#106)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| MMLU-Pro | 81.9% | 78.6% |
| Vectara Hallucination Rate | 17.9% | 9.3% |
| GPQA (HELM) | 65.3% | 63% |
| LMArena Expert | 1365 | 1417 |
| Confabulations | 20.4% | — |
Multilingual Qwen3-Next 80B-A3B Instruct leads
Kimi K2 (Jul 2025): 49.6 (#130), Qwen3-Next 80B-A3B Instruct: 52.1 (#93)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Non-English | 1372 | 1407 |
| LMArena Chinese | 1415 | 1460 |
| LMArena French | 1379 | 1413 |
| LMArena German | 1387 | 1417 |
| LMArena Japanese | 1349 | 1395 |
| LMArena Korean | 1325 | 1364 |
| LMArena Russian | 1385 | 1404 |
| LMArena Spanish | 1386 | 1435 |
Instruction Following Too close to call
Kimi K2 (Jul 2025): 71.1 (#156), Qwen3-Next 80B-A3B Instruct: 70.8 (#159)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| IFEval | 85% | 81% |
| LMArena Instruction Following | 1348 | 1389 |
Long Context Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 41.2 (#145), Qwen3-Next 80B-A3B Instruct: 37.0 (#223)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| Fiction.LiveBench | 66.7% | 55.6% |
| LMArena Longer Query | 1353 | 1403 |
| CL-bench | 17.6% | — |
Writing & Preference Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 62.3 (#78), Qwen3-Next 80B-A3B Instruct: 58.0 (#121)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Text | 1380 | 1417 |
| LMArena Creative Writing | 1350 | 1334 |
| WildBench | 86.2% | 80.7% |
| LMArena Multi-Turn | 1371 | 1416 |
| Short-Story Creative Writing | 85.6% | — |
| EQ-Bench Creative Writing | 1666 | — |
Frequently asked questions
Is Kimi K2 (Jul 2025) better than Qwen3-Next 80B-A3B Instruct?
Qwen3-Next 80B-A3B Instruct is the stronger model overall, scoring 43.0 to 41.2 on the Noometry Index.
Which is cheaper, Kimi K2 (Jul 2025) or Qwen3-Next 80B-A3B Instruct?
Qwen3-Next 80B-A3B Instruct is cheaper. It lists at $0.50 per million input tokens and $2 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is Kimi K2 (Jul 2025) or Qwen3-Next 80B-A3B Instruct better for coding?
They score almost the same on coding (42.4 vs 42.5); test both on your own repository before choosing.
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-Next 80B-A3B Instruct share?
25 benchmarks have published results for both models. Kimi K2 (Jul 2025) has 42 scored results on Noometry and Qwen3-Next 80B-A3B Instruct has 25.