Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
GLM-5V-Turbo
~58
1/8 categoriesQwen2.5-72B
61
Winner · 0/8 categoriesGLM-5V-Turbo· Qwen2.5-72B
Pick Qwen2.5-72B if you want the stronger benchmark profile. GLM-5V-Turbo only becomes the better choice if agentic is the priority or you need the larger 200K context window.
Qwen2.5-72B has the cleaner overall profile here, landing at 61 versus 58. It is a real lead, but still close enough that category-level strengths matter more than the headline number.
GLM-5V-Turbo is also the more expensive model on tokens at $1.20 input / $4.00 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Qwen2.5-72B. That is roughly Infinityx on output cost alone. GLM-5V-Turbo gives you the larger context window at 200K, compared with 128K for Qwen2.5-72B.
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 | GLM-5V-Turbo | Qwen2.5-72B |
|---|---|---|
| AgenticGLM-5V-Turbo wins | ||
| BrowseComp | 51.9% | 64% |
| OSWorld-Verified | 62.3% | 55% |
| BrowseComp-VL | 51.9% | — |
| OSWorld | 62.3% | — |
| AndroidWorld | 75.7% | — |
| WebVoyager | 88.5% | — |
| Terminal-Bench 2.0 | — | 56% |
| Coding | ||
| HumanEval | — | 75% |
| SWE-bench Verified | — | 46% |
| LiveCodeBench | — | 40% |
| SWE-bench Pro | — | 47% |
| Multimodal & Grounded | ||
| Design2Code | 94.8% | — |
| Flame-VLM-Code | 93.8% | — |
| Vision2Web | 31.0% | — |
| ImageMining | 30.7% | — |
| MMSearch | 72.9% | — |
| MMSearch-Plus | 30.0% | — |
| SimpleVQA | 78.2% | — |
| Facts-VLM | 58.6% | — |
| V* | 89.0% | — |
| MMMU-Pro | — | 64% |
| OfficeQA Pro | — | 70% |
| Reasoning | ||
| MuSR | — | 78% |
| BBH | — | 81% |
| MRCRv2 | — | 71% |
| Knowledge | ||
| MMLU | — | 83% |
| GPQA | — | 82% |
| SuperGPQA | — | 80% |
| MMLU-Pro | — | 75% |
| HLE | — | 11% |
| FrontierScience | — | 70% |
| SimpleQA | — | 80% |
| Instruction Following | ||
| IFEval | — | 85% |
| Multilingual | ||
| MGSM | — | 84% |
| MMLU-ProX | — | 79% |
| Mathematics | ||
| AIME 2023 | — | 84% |
| AIME 2024 | — | 86% |
| AIME 2025 | — | 85% |
| HMMT Feb 2023 | — | 80% |
| HMMT Feb 2024 | — | 82% |
| HMMT Feb 2025 | — | 81% |
| BRUMO 2025 | — | 83% |
| MATH-500 | — | 84% |
Qwen2.5-72B is ahead overall, 61 to 58. The biggest single separator in this matchup is BrowseComp, where the scores are 51.9% and 64%.
GLM-5V-Turbo has the edge for agentic tasks in this comparison, averaging 58 versus 57.7. Inside this category, BrowseComp 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.