Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
GPT-5.4 mini
68
Winner · 2/8 categoriesQwen3.5-35B-A3B
67
4/8 categoriesGPT-5.4 mini· Qwen3.5-35B-A3B
Pick GPT-5.4 mini if you want the stronger benchmark profile. Qwen3.5-35B-A3B only becomes the better choice if knowledge is the priority or you want the cheaper token bill.
GPT-5.4 mini finishes one point ahead overall, 68 to 67. That is enough to call, but not enough to treat as a blowout. This matchup comes down to a few meaningful edges rather than one model dominating the board.
GPT-5.4 mini's sharpest advantage is in agentic, where it averages 65.6 against 50.5. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 60% to 40.5%. Qwen3.5-35B-A3B does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.
GPT-5.4 mini is also the more expensive model on tokens at $0.75 input / $4.50 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Qwen3.5-35B-A3B. That is roughly Infinityx on output cost alone. GPT-5.4 mini gives you the larger context window at 400K, compared with 262K for Qwen3.5-35B-A3B.
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 | GPT-5.4 mini | Qwen3.5-35B-A3B |
|---|---|---|
| AgenticGPT-5.4 mini wins | ||
| Terminal-Bench 2.0 | 60% | 40.5% |
| OSWorld-Verified | 72.1% | 54.5% |
| MCP Atlas | 57.7% | — |
| Toolathlon | 42.9% | — |
| tau2-bench | 93.4% | 81.2% |
| BrowseComp | — | 61% |
| CodingQwen3.5-35B-A3B wins | ||
| SWE-bench Pro | 54.4% | — |
| SWE-bench Verified | — | 69.2% |
| LiveCodeBench | — | 74.6% |
| Multimodal & GroundedGPT-5.4 mini wins | ||
| MMMU-Pro | 76.6% | 75.1% |
| MMMU-Pro w/ Python | 78% | — |
| OmniDocBench 1.5 | 0.1263 | — |
| ReasoningQwen3.5-35B-A3B wins | ||
| MRCRv2 | 40.7% | — |
| MRCR v2 64K-128K | 47.7% | — |
| MRCR v2 128K-256K | 33.6% | — |
| Graphwalks BFS 128K | 76.3% | — |
| Graphwalks Parents 128K | 71.5% | — |
| LongBench v2 | — | 59% |
| KnowledgeQwen3.5-35B-A3B wins | ||
| GPQA | 88% | 84.2% |
| HLE | 41.5% | — |
| HLE w/o tools | 28.2% | — |
| MMLU-Pro | — | 85.3% |
| SuperGPQA | — | 63.4% |
| Instruction FollowingQwen3.5-35B-A3B wins | ||
| IFEval | 87.4% | 91.9% |
| Multilingual | ||
| MMLU-ProX | — | 81% |
| Mathematics | ||
| MATH-500 | 97.4% | — |
GPT-5.4 mini is ahead overall, 68 to 67. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 60% and 40.5%.
Qwen3.5-35B-A3B has the edge for knowledge tasks in this comparison, averaging 79.3 versus 57.4. 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 54.4. GPT-5.4 mini stays close enough that the answer can still flip depending on your workload.
Qwen3.5-35B-A3B has the edge for reasoning in this comparison, averaging 59 versus 40.7. GPT-5.4 mini stays close enough that the answer can still flip depending on your workload.
GPT-5.4 mini has the edge for agentic tasks in this comparison, averaging 65.6 versus 50.5. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
GPT-5.4 mini has the edge for multimodal and grounded tasks in this comparison, averaging 76.6 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 87.4. 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.