Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Sibling matchup inside the GLM-5 family.
GLM-5 (Reasoning)
82
Winner · 1/8 categoriesGLM-5V-Turbo
~58
0/8 categoriesGLM-5 (Reasoning)· GLM-5V-Turbo
GLM-5 (Reasoning) makes more sense if agentic is the priority or you want the cheaper token bill, while GLM-5V-Turbo is the cleaner fit if you would rather avoid the extra latency and token burn of a reasoning model.
GLM-5 (Reasoning) and GLM-5V-Turbo sit in the same GLM-5 family. This page is less about two unrelated model lineages and more about how the siblings trade off on benchmark shape, token costs, and practical limits like context window.
GLM-5 (Reasoning) is clearly ahead on the aggregate, 82 to 58. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GLM-5 (Reasoning)'s sharpest advantage is in agentic, where it averages 78.3 against 58. The single biggest benchmark swing on the page is BrowseComp, 80% to 51.9%.
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 GLM-5 (Reasoning). That is roughly Infinityx on output cost alone. GLM-5 (Reasoning) is the reasoning model in the pair, while GLM-5V-Turbo 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.
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-5 (Reasoning) | GLM-5V-Turbo |
|---|---|---|
| AgenticGLM-5 (Reasoning) wins | ||
| Terminal-Bench 2.0 | 81% | — |
| BrowseComp | 80% | 51.9% |
| OSWorld-Verified | 74% | 62.3% |
| BrowseComp-VL | — | 51.9% |
| OSWorld | — | 62.3% |
| AndroidWorld | — | 75.7% |
| WebVoyager | — | 88.5% |
| Coding | ||
| HumanEval | 88% | — |
| SWE-bench Verified | 62% | — |
| LiveCodeBench | 58% | — |
| SWE-bench Pro | 67% | — |
| Multimodal & Grounded | ||
| MMMU-Pro | 74% | — |
| OfficeQA Pro | 84% | — |
| 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% |
| Reasoning | ||
| MuSR | 90% | — |
| BBH | 91% | — |
| LongBench v2 | 86% | — |
| MRCRv2 | 87% | — |
| Knowledge | ||
| MMLU | 96% | — |
| GPQA | 94% | — |
| SuperGPQA | 92% | — |
| MMLU-Pro | 81% | — |
| HLE | 29% | — |
| FrontierScience | 83% | — |
| SimpleQA | 92% | — |
| Instruction Following | ||
| IFEval | 92% | — |
| Multilingual | ||
| MGSM | 89% | — |
| MMLU-ProX | 85% | — |
| Mathematics | ||
| AIME 2023 | 98% | — |
| AIME 2024 | 99% | — |
| AIME 2025 | 98% | — |
| HMMT Feb 2023 | 94% | — |
| HMMT Feb 2024 | 96% | — |
| HMMT Feb 2025 | 95% | — |
| BRUMO 2025 | 96% | — |
| MATH-500 | 92% | — |
GLM-5 (Reasoning) and GLM-5V-Turbo are sibling variants in the GLM-5 family, so the right pick depends on whether you value the better benchmark line, cheaper tokens, or the larger context window. GLM-5 (Reasoning) is ahead overall 82 to 58.
GLM-5 (Reasoning) has the edge for agentic tasks in this comparison, averaging 78.3 versus 58. 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.