Model comparison
Kimi K2 (Jul 2025) vs Qwen3-30B-A3B
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 38.9 on the Noometry Index. Qwen3-30B-A3B costs 4.7× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. Kimi K2 (Jul 2025) scores higher in 7 categories and Qwen3-30B-A3B in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where Kimi K2 (Jul 2025) leads 41.2 to 31.0.
- The biggest single-benchmark swing is Fiction.LiveBench: 66.7% for Kimi K2 (Jul 2025) and 40.6% for Qwen3-30B-A3B.
- Qwen3-30B-A3B is cheaper at $0.12 / $0.50 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 41K.
Side by side
| Kimi K2 (Jul 2025) | Qwen3-30B-A3B | |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 41.2 | 38.9 |
| Released | 2025-07-12 | 2025-04-28 |
| Weights | Open | Open |
| Context window | 262K | 41K |
| Max output | 262K | 16K |
| Input $ / M tokens | $0.57 | $0.12 |
| Output $ / M tokens | $2.30 | $0.50 |
| Results tracked | 42 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 42.4 (#102), Qwen3-30B-A3B: 37.5 (#194)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-30B-A3B |
|---|---|---|
| WeirdML | 42.8% | 29.8% |
| LMArena Coding | 1399 | 1416 |
| SWE-bench Verified (bash only) | 63.4% | — |
| Aider Polyglot | 59.1% | — |
| SciCode | — | 33.3% |
| GSO | 4.9% | — |
| ALE-Bench | 597.5 | — |
Agentic & Tool Use Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 32.4 (#64), Qwen3-30B-A3B: 29.8 (#82)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 59.1% | 41.4% |
| Terminal-Bench | 35.7% | — |
| METR Time Horizons | 59.2% | — |
Reasoning Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 23.3 (#179), Qwen3-30B-A3B: 22.2 (#204)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-30B-A3B |
|---|---|---|
| Kagi LLM Benchmark | 64.4% | 54.9% |
| LMArena Hard Prompts | 1384 | 1398 |
| Epoch Capabilities Index | 146.01 | 139.63 |
| SimpleBench | 26.3% | — |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 8% |
| DTBench | — | 69.3% |
| LMCA | — | 22.4% |
| ForecastBench | 60.2 | — |
Math Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 42.7 (#83), Qwen3-30B-A3B: 37.4 (#157)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-30B-A3B |
|---|---|---|
| LMArena Math | 1397 | 1394 |
| MathArena Final-Answer Competitions | — | 47.8% |
| OTIS Mock AIME 2024-2025 | — | 70.3% |
| Omni-MATH | 65.4% | — |
| FrontierMath (Feb 2025 set) | 21.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Qwen3-30B-A3B leads
Kimi K2 (Jul 2025): 37.3 (#157), Qwen3-30B-A3B: 41.8 (#105)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-30B-A3B |
|---|---|---|
| Confabulations | 20.4% | 12.3% |
| LMArena Expert | 1365 | 1396 |
| GPQA Diamond | — | 70.1% |
| MMLU-Pro | 81.9% | — |
| Vectara Hallucination Rate | 17.9% | — |
| GPQA (HELM) | 65.3% | — |
Multilingual Too close to call
Kimi K2 (Jul 2025): 49.6 (#130), Qwen3-30B-A3B: 49.5 (#132)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1372 | 1372 |
| LMArena Chinese | 1415 | 1433 |
| LMArena French | 1379 | 1418 |
| LMArena German | 1387 | 1380 |
| LMArena Japanese | 1349 | 1337 |
| LMArena Korean | 1325 | 1331 |
| LMArena Russian | 1385 | 1370 |
| LMArena Spanish | 1386 | 1404 |
Instruction Following Too close to call
Kimi K2 (Jul 2025): 71.1 (#156), Qwen3-30B-A3B: 72.0 (#142)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1348 | 1363 |
| IFEval | 85% | — |
Long Context Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 41.2 (#145), Qwen3-30B-A3B: 31.0 (#283)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-30B-A3B |
|---|---|---|
| Fiction.LiveBench | 66.7% | 40.6% |
| LMArena Longer Query | 1353 | 1379 |
| CL-bench | 17.6% | — |
Writing & Preference Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 62.3 (#78), Qwen3-30B-A3B: 55.6 (#143)
| Benchmark | Kimi K2 (Jul 2025) | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1380 | 1384 |
| LMArena Creative Writing | 1350 | 1317 |
| Short-Story Creative Writing | 85.6% | 75.3% |
| LMArena Multi-Turn | 1371 | 1378 |
| EQ-Bench Creative Writing | 1666 | — |
| WildBench | 86.2% | — |
Frequently asked questions
Is Kimi K2 (Jul 2025) better than Qwen3-30B-A3B?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 38.9 on the Noometry Index. Qwen3-30B-A3B costs 4.7× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.
Which is cheaper, Kimi K2 (Jul 2025) or Qwen3-30B-A3B?
Qwen3-30B-A3B is cheaper. It lists at $0.12 per million input tokens and $0.50 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is Kimi K2 (Jul 2025) or Qwen3-30B-A3B better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 37.5 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 41K.
How many benchmarks do Kimi K2 (Jul 2025) and Qwen3-30B-A3B share?
24 benchmarks have published results for both models. Kimi K2 (Jul 2025) has 42 scored results on Noometry and Qwen3-30B-A3B has 32.