Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Nemotron-4 15B
46
0/8 categoriesQwen3.5-122B-A10B
71
Winner · 6/8 categoriesNemotron-4 15B· Qwen3.5-122B-A10B
Pick Qwen3.5-122B-A10B if you want the stronger benchmark profile. Nemotron-4 15B only becomes the better choice if you would rather avoid the extra latency and token burn of a reasoning model.
Qwen3.5-122B-A10B is clearly ahead on the aggregate, 71 to 46. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Qwen3.5-122B-A10B's sharpest advantage is in coding, where it averages 76.3 against 27.2. The single biggest benchmark swing on the page is LiveCodeBench, 22% to 78.9%.
Qwen3.5-122B-A10B is the reasoning model in the pair, while Nemotron-4 15B 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. Qwen3.5-122B-A10B gives you the larger context window at 262K, compared with 32K for Nemotron-4 15B.
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 | Nemotron-4 15B | Qwen3.5-122B-A10B |
|---|---|---|
| AgenticQwen3.5-122B-A10B wins | ||
| Terminal-Bench 2.0 | 37% | 49.4% |
| BrowseComp | 47% | 63.8% |
| OSWorld-Verified | 42% | 58% |
| tau2-bench | — | 79.5% |
| CodingQwen3.5-122B-A10B wins | ||
| HumanEval | 46% | — |
| SWE-bench Verified | 31% | 72% |
| LiveCodeBench | 22% | 78.9% |
| SWE-bench Pro | 30% | — |
| Multimodal & GroundedQwen3.5-122B-A10B wins | ||
| MMMU-Pro | 46% | 76.9% |
| OfficeQA Pro | 54% | — |
| ReasoningQwen3.5-122B-A10B wins | ||
| MuSR | 50% | — |
| BBH | 73% | — |
| LongBench v2 | 52% | 60.2% |
| MRCRv2 | 51% | — |
| KnowledgeQwen3.5-122B-A10B wins | ||
| MMLU | 54% | — |
| GPQA | 53% | 86.6% |
| SuperGPQA | 51% | 67.1% |
| HLE | 5% | — |
| FrontierScience | 50% | — |
| SimpleQA | 52% | — |
| MMLU-Pro | — | 86.7% |
| Instruction Following | ||
| IFEval | — | 93.4% |
| MultilingualQwen3.5-122B-A10B wins | ||
| MGSM | 75% | — |
| MMLU-ProX | 71% | 82.2% |
| Mathematics | ||
| AIME 2023 | 54% | — |
| AIME 2024 | 56% | — |
| AIME 2025 | 55% | — |
| HMMT Feb 2023 | 50% | — |
| HMMT Feb 2024 | 52% | — |
| HMMT Feb 2025 | 51% | — |
| BRUMO 2025 | 53% | — |
Qwen3.5-122B-A10B is ahead overall, 71 to 46. The biggest single separator in this matchup is LiveCodeBench, where the scores are 22% and 78.9%.
Qwen3.5-122B-A10B has the edge for knowledge tasks in this comparison, averaging 81.6 versus 37.7. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Qwen3.5-122B-A10B has the edge for coding in this comparison, averaging 76.3 versus 27.2. Inside this category, LiveCodeBench is the benchmark that creates the most daylight between them.
Qwen3.5-122B-A10B has the edge for reasoning in this comparison, averaging 60.2 versus 51.1. Inside this category, LongBench v2 is the benchmark that creates the most daylight between them.
Qwen3.5-122B-A10B has the edge for agentic tasks in this comparison, averaging 56 versus 41.3. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.
Qwen3.5-122B-A10B has the edge for multimodal and grounded tasks in this comparison, averaging 76.9 versus 49.6. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
Qwen3.5-122B-A10B has the edge for multilingual tasks in this comparison, averaging 82.2 versus 72.4. 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.