Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
GPT-5.3 Codex
85
Winner · 5/8 categoriesQwen3.5-35B-A3B
67
2/8 categoriesGPT-5.3 Codex· Qwen3.5-35B-A3B
Pick GPT-5.3 Codex if you want the stronger benchmark profile. Qwen3.5-35B-A3B only becomes the better choice if coding is the priority or you want the cheaper token bill.
GPT-5.3 Codex is clearly ahead on the aggregate, 85 to 67. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GPT-5.3 Codex's sharpest advantage is in reasoning, where it averages 92.5 against 59. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 77.3% to 40.5%. Qwen3.5-35B-A3B does hit back in coding, so the answer changes if that is the part of the workload you care about most.
GPT-5.3 Codex is also the more expensive model on tokens at $2.50 input / $10.00 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-5.3 Codex 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-5.3 Codex | Qwen3.5-35B-A3B |
|---|---|---|
| AgenticGPT-5.3 Codex wins | ||
| Terminal-Bench 2.0 | 77.3% | 40.5% |
| BrowseComp | 88% | 61% |
| OSWorld-Verified | 64.7% | 54.5% |
| tau2-bench | — | 81.2% |
| CodingQwen3.5-35B-A3B wins | ||
| SWE-bench Verified | 85% | 69.2% |
| LiveCodeBench | 85% | 74.6% |
| SWE-bench Pro | 56.8% | — |
| SWE-Rebench | 58.2% | — |
| React Native Evals | 80.9% | — |
| Multimodal & GroundedGPT-5.3 Codex wins | ||
| MMMU-Pro | 89% | 75.1% |
| OfficeQA Pro | 94% | — |
| ReasoningGPT-5.3 Codex wins | ||
| BBH | 98% | — |
| LongBench v2 | 92% | 59% |
| MRCRv2 | 93% | — |
| KnowledgeQwen3.5-35B-A3B wins | ||
| MMLU-Pro | 90% | 85.3% |
| HLE | 44% | — |
| FrontierScience | 90% | — |
| SimpleQA | 95% | — |
| SuperGPQA | — | 63.4% |
| GPQA | — | 84.2% |
| Instruction FollowingGPT-5.3 Codex wins | ||
| IFEval | 93% | 91.9% |
| MultilingualGPT-5.3 Codex wins | ||
| MGSM | 96% | — |
| MMLU-ProX | 91% | 81% |
| Mathematics | ||
| AIME 2023 | 99% | — |
| AIME 2024 | 99% | — |
| AIME 2025 | 98% | — |
| HMMT Feb 2023 | 95% | — |
| HMMT Feb 2024 | 97% | — |
| HMMT Feb 2025 | 96% | — |
| BRUMO 2025 | 96% | — |
| MATH-500 | 99% | — |
GPT-5.3 Codex is ahead overall, 85 to 67. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 77.3% and 40.5%.
Qwen3.5-35B-A3B has the edge for knowledge tasks in this comparison, averaging 79.3 versus 76.9. Inside this category, MMLU-Pro 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 68.6. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for reasoning in this comparison, averaging 92.5 versus 59. Inside this category, LongBench v2 is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for agentic tasks in this comparison, averaging 75.6 versus 50.5. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for multimodal and grounded tasks in this comparison, averaging 91.3 versus 75.1. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for instruction following in this comparison, averaging 93 versus 91.9. Inside this category, IFEval is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for multilingual tasks in this comparison, averaging 92.8 versus 81. 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.