Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Kimi K2
60
1/8 categoriesQwen3.5-35B-A3B
68
Winner · 3/8 categoriesKimi K2· Qwen3.5-35B-A3B
Pick Qwen3.5-35B-A3B if you want the stronger benchmark profile. Kimi K2 only becomes the better choice if agentic is the priority or you would rather avoid the extra latency and token burn of a reasoning model.
Qwen3.5-35B-A3B is clearly ahead on the aggregate, 68 to 60. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Qwen3.5-35B-A3B's sharpest advantage is in knowledge, where it averages 79.3 against 64. The single biggest benchmark swing on the page is LiveCodeBench, 53.7% to 74.6%. Kimi K2 does hit back in agentic, so the answer changes if that is the part of the workload you care about most.
Qwen3.5-35B-A3B is the reasoning model in the pair, while Kimi K2 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Qwen3.5-35B-A3B gives you the larger context window at 262K, compared with 128K for Kimi K2.
BenchLM keeps the benchmark table and the operator tradeoffs on the same page so a better score does not hide a materially slower, pricier, or smaller-context model.
Runtime metrics show N/A when BenchLM does not have a sourced snapshot for that exact model. The scoring rules and freshness policy are documented on the methodology page.
| Benchmark | Kimi K2 | Qwen3.5-35B-A3B |
|---|---|---|
| AgenticKimi K2 wins | ||
| Terminal-Bench 2.0 | 47.1% | 40.5% |
| BrowseComp | 60.2% | 61% |
| tau2-bench | 66.1% | 81.2% |
| OSWorld-Verified | — | 54.5% |
| CodingQwen3.5-35B-A3B wins | ||
| SWE-bench Verified | 65.8% | 69.2% |
| LiveCodeBench | 53.7% | 74.6% |
| Multimodal & Grounded | ||
| MMMU-Pro | — | 75.1% |
| Reasoning | ||
| hle | 44.9% | — |
| LongBench v2 | — | 59% |
| KnowledgeQwen3.5-35B-A3B wins | ||
| MMLU | 89.5% | — |
| GPQA | 75.1% | 84.2% |
| SuperGPQA | 57.2% | 63.4% |
| MMLU-Pro | 81.1% | 85.3% |
| SimpleQA | 31% | — |
| Instruction FollowingQwen3.5-35B-A3B wins | ||
| IFEval | 89.8% | 91.9% |
| Multilingual | ||
| sweMultilingual | 61.1% | — |
| MMLU-ProX | — | 81% |
| Mathematics | ||
| AIME 2024 | 69.6% | — |
| AIME 2025 | 49.5% | — |
| MATH-500 | 97.4% | — |
| HMMT Feb 2025 | 38.8% | — |
Qwen3.5-35B-A3B is ahead overall, 68 to 60. The biggest single separator in this matchup is LiveCodeBench, where the scores are 53.7% and 74.6%.
Qwen3.5-35B-A3B has the edge for knowledge tasks in this comparison, averaging 79.3 versus 64. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Qwen3.5-35B-A3B has the edge for coding in this comparison, averaging 72.6 versus 58.2. Inside this category, LiveCodeBench is the benchmark that creates the most daylight between them.
Kimi K2 has the edge for agentic tasks in this comparison, averaging 52.1 versus 50.5. Inside this category, tau2-bench is the benchmark that creates the most daylight between them.
Qwen3.5-35B-A3B has the edge for instruction following in this comparison, averaging 91.9 versus 89.8. Inside this category, IFEval is the benchmark that creates the most daylight between them.
Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.
Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.