Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
MiMo-V2-Flash
67
3/8 categoriesQwen3.5-35B-A3B
67
4/8 categoriesMiMo-V2-Flash· Qwen3.5-35B-A3B
Treat this as a split decision. MiMo-V2-Flash makes more sense if agentic is the priority; Qwen3.5-35B-A3B is the better fit if knowledge is the priority or you need the larger 262K context window.
MiMo-V2-Flash and Qwen3.5-35B-A3B finish on the same overall score, so this is less about a single winner and more about where the edge shows up. The headline says tie; the benchmark table is where the real choice happens.
Qwen3.5-35B-A3B gives you the larger context window at 262K, compared with 256K for MiMo-V2-Flash.
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 | MiMo-V2-Flash | Qwen3.5-35B-A3B |
|---|---|---|
| AgenticMiMo-V2-Flash wins | ||
| Terminal-Bench 2.0 | 63% | 40.5% |
| BrowseComp | 65% | 61% |
| OSWorld-Verified | 58% | 54.5% |
| tau2-bench | — | 81.2% |
| CodingQwen3.5-35B-A3B wins | ||
| HumanEval | 84.8% | — |
| SWE-bench Verified | 73.4% | 69.2% |
| LiveCodeBench | 80.6% | 74.6% |
| SWE-bench Pro | 52% | — |
| SWE Multilingual | 71.7% | — |
| Multimodal & GroundedMiMo-V2-Flash wins | ||
| MMMU-Pro | 78% | 75.1% |
| OfficeQA Pro | 73% | — |
| ReasoningMiMo-V2-Flash wins | ||
| MuSR | 74% | — |
| BBH | 85% | — |
| LongBench v2 | 60.6% | 59% |
| MRCRv2 | 73% | — |
| KnowledgeQwen3.5-35B-A3B wins | ||
| MMLU | 86.7% | — |
| GPQA | 83.7% | 84.2% |
| SuperGPQA | 76% | 63.4% |
| MMLU-Pro | 84.9% | 85.3% |
| HLE | 14% | — |
| FrontierScience | 71% | — |
| SimpleQA | 76% | — |
| Instruction FollowingQwen3.5-35B-A3B wins | ||
| IFEval | 84% | 91.9% |
| MultilingualQwen3.5-35B-A3B wins | ||
| MGSM | 83% | — |
| MMLU-ProX | 77% | 81% |
| Mathematics | ||
| AIME 2023 | 79% | — |
| AIME 2024 | 81% | — |
| AIME 2025 | 94.1% | — |
| HMMT Feb 2023 | 75% | — |
| HMMT Feb 2024 | 77% | — |
| HMMT Feb 2025 | 76% | — |
| BRUMO 2025 | 78% | — |
| MATH-500 | 90% | — |
MiMo-V2-Flash and Qwen3.5-35B-A3B are tied on overall score, so the right pick depends on which category matters most for your use case.
Qwen3.5-35B-A3B has the edge for knowledge tasks in this comparison, averaging 79.3 versus 63.7. Inside this category, SuperGPQA 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 67.9. Inside this category, LiveCodeBench is the benchmark that creates the most daylight between them.
MiMo-V2-Flash has the edge for reasoning in this comparison, averaging 68.3 versus 59. Inside this category, LongBench v2 is the benchmark that creates the most daylight between them.
MiMo-V2-Flash has the edge for agentic tasks in this comparison, averaging 61.8 versus 50.5. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
MiMo-V2-Flash has the edge for multimodal and grounded tasks in this comparison, averaging 75.8 versus 75.1. Inside this category, MMMU-Pro 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 84. Inside this category, IFEval is the benchmark that creates the most daylight between them.
Qwen3.5-35B-A3B has the edge for multilingual tasks in this comparison, averaging 81 versus 79.1. Inside this category, MMLU-ProX 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.