Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
GLM-5V-Turbo
~58
1/8 categoriesMistral Large 3
58
0/8 categoriesGLM-5V-Turbo· Mistral Large 3
Treat this as a split decision. GLM-5V-Turbo makes more sense if agentic is the priority or you want the cheaper token bill; Mistral Large 3 is the better fit if its strengths line up with your actual workload.
GLM-5V-Turbo and Mistral Large 3 finish on the same overall score, so this is less about a single winner and more about where the edge shows up. The headline says tie; the benchmark table is where the real choice happens.
Mistral Large 3 is also the more expensive model on tokens at $2.00 input / $6.00 output per 1M tokens, versus $1.20 input / $4.00 output per 1M tokens for GLM-5V-Turbo. GLM-5V-Turbo gives you the larger context window at 200K, compared with 128K for Mistral Large 3.
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 | Mistral Large 3 |
|---|---|---|
| AgenticGLM-5V-Turbo wins | ||
| BrowseComp | 51.9% | 58% |
| OSWorld-Verified | 62.3% | 49% |
| BrowseComp-VL | 51.9% | — |
| OSWorld | 62.3% | — |
| AndroidWorld | 75.7% | — |
| WebVoyager | 88.5% | — |
| Terminal-Bench 2.0 | — | 52% |
| Coding | ||
| HumanEval | — | 92.3% |
| SWE-bench Verified | — | 45% |
| LiveCodeBench | — | 39% |
| SWE-bench Pro | — | 42% |
| 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 | — | 75% |
| OfficeQA Pro | — | 76% |
| Reasoning | ||
| MuSR | — | 71% |
| BBH | — | 81% |
| LongBench v2 | — | 67% |
| MRCRv2 | — | 67% |
| Knowledge | ||
| MMLU | — | 85.5% |
| GPQA | — | 43.9% |
| SuperGPQA | — | 73% |
| MMLU-Pro | — | 73.1% |
| HLE | — | 12% |
| FrontierScience | — | 67% |
| SimpleQA | — | 24% |
| Instruction Following | ||
| IFEval | — | 89.4% |
| Multilingual | ||
| MGSM | — | 82% |
| MMLU-ProX | — | 77% |
| Mathematics | ||
| AIME 2023 | — | 76% |
| AIME 2024 | — | 78% |
| AIME 2025 | — | 77% |
| HMMT Feb 2023 | — | 72% |
| HMMT Feb 2024 | — | 74% |
| HMMT Feb 2025 | — | 73% |
| BRUMO 2025 | — | 75% |
| MATH-500 | — | 93.6% |
GLM-5V-Turbo and Mistral Large 3 are tied on overall score, so the right pick depends on which category matters most for your use case.
GLM-5V-Turbo has the edge for agentic tasks in this comparison, averaging 58 versus 52.5. Inside this category, OSWorld-Verified 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.