Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Gemma 4 31B
73
1/8 categoriesGPT-5.2-Codex
82
Winner · 3/8 categoriesGemma 4 31B· GPT-5.2-Codex
Pick GPT-5.2-Codex if you want the stronger benchmark profile. Gemma 4 31B only becomes the better choice if coding is the priority or you want the cheaper token bill.
GPT-5.2-Codex is clearly ahead on the aggregate, 82 to 73. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GPT-5.2-Codex's sharpest advantage is in reasoning, where it averages 91.1 against 66.4. The single biggest benchmark swing on the page is MRCRv2, 66.4% to 91%. Gemma 4 31B does hit back in coding, so the answer changes if that is the part of the workload you care about most.
GPT-5.2-Codex is also the more expensive model on tokens at $2.00 input / $8.00 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Gemma 4 31B. That is roughly Infinityx on output cost alone. GPT-5.2-Codex gives you the larger context window at 400K, compared with 256K for Gemma 4 31B.
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 | Gemma 4 31B | GPT-5.2-Codex |
|---|---|---|
| Agentic | ||
| Terminal-Bench 2.0 | — | 90% |
| BrowseComp | — | 85% |
| OSWorld-Verified | — | 85% |
| CodingGemma 4 31B wins | ||
| LiveCodeBench | 80% | 66% |
| HumanEval | — | 95% |
| SWE-bench Verified | — | 76% |
| SWE-bench Pro | — | 86% |
| SWE-Rebench | — | 56.8% |
| Multimodal & GroundedGPT-5.2-Codex wins | ||
| MMMU-Pro | 76.9% | 84% |
| OfficeQA Pro | — | 92% |
| ReasoningGPT-5.2-Codex wins | ||
| BBH | 74.4% | 90% |
| MRCRv2 | 66.4% | 91% |
| MuSR | — | 93% |
| LongBench v2 | — | 90% |
| KnowledgeGPT-5.2-Codex wins | ||
| GPQA | 84.3% | 97% |
| MMLU-Pro | 85.2% | 80% |
| HLE | 26.5% | 26% |
| HLE w/o tools | 19.5% | — |
| MMLU | — | 99% |
| SuperGPQA | — | 95% |
| FrontierScience | — | 86% |
| SimpleQA | — | 95% |
| Instruction Following | ||
| IFEval | — | 92% |
| Multilingual | ||
| MGSM | — | 91% |
| MMLU-ProX | — | 87% |
| 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 | — | 94% |
GPT-5.2-Codex is ahead overall, 82 to 73. The biggest single separator in this matchup is MRCRv2, where the scores are 66.4% and 91%.
GPT-5.2-Codex has the edge for knowledge tasks in this comparison, averaging 74.5 versus 61.3. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Gemma 4 31B has the edge for coding in this comparison, averaging 80 versus 69.3. Inside this category, LiveCodeBench is the benchmark that creates the most daylight between them.
GPT-5.2-Codex has the edge for reasoning in this comparison, averaging 91.1 versus 66.4. Inside this category, MRCRv2 is the benchmark that creates the most daylight between them.
GPT-5.2-Codex has the edge for multimodal and grounded tasks in this comparison, averaging 87.6 versus 76.9. Inside this category, MMMU-Pro 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.