MATH
Competition maths problems (Phi-4-mini card)
math
Score in %
Definition
- Version
- not in the notes
- Category
- math
- Maintainer
- not in the notes
- Status
- not in the notes
- Home page
- not in the notes
Sources
- allenai/Olmo-3-32B-Think , Hugging Face huggingface.co https://huggingface.co/allenai/Olmo-3-32B-Think
- huggingface.co/microsoft/Phi-4-mini-instruct https://huggingface.co/microsoft/Phi-4-mini-instruct
Scores at a Glance
Score in %
Higher is better for this unit.
1
Olmo 3 32B ThinkAllen AI
Phi-4-mini-instructMicrosoft AI
GLM-4 (0520)Z.ai (Zhipu)
Mixtral 8x22B InstructMistral AImaj@4
Qwen1.5-72BAlibaba (Qwen)
Olmo 3 32B ThinkAllen AI96.1 %
vendor-reported2Phi-4-mini-instructMicrosoft AI64 %
vendor-reported3GLM-4 (0520)Z.ai (Zhipu)61.3 %
vendor-reported4Mixtral 8x22B InstructMistral AImaj@444.6 %
vendor-reported5Qwen1.5-72BAlibaba (Qwen)34.1 %
vendor-reported0255075100
Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.
Scores
| Model | Score | Who ran it | Setting | View or population | Date | Source |
|---|---|---|---|---|---|---|
| Olmo 3 32B Think | 96.1 % | vendor-reported | – | – | – | huggingface.co |
| Olmo 3.1 32B Instruct | 93.4 % | vendor-reported | – | Olmo 3.1 32B Instruct model card | Dec 2025 | huggingface.co |
| Phi-4-mini-instruct | 64 % | vendor-reported | – | – | – | huggingface.co |
| GLM-4 (0520) | 61.3 % | vendor-reported | – | – | 18 Jun 2024 | arxiv.org |
| OLMo 2 32B (OLMo-2-0325-32B-Instruct) | 49.7 % | vendor-reported | – | OLMo-2-0325-32B-Instruct model card | Mar 2025 | huggingface.co |
| Mixtral 8x22B Instruct | 44.6 % | vendor-reported | maj@4 | – | 17 Apr 2024 | mistral.ai |
| Qwen1.5-72B | 34.1 % | vendor-reported | – | – | 4 Feb 2024 | qwenlm.github.io |
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "MATH", https://aurion.semai.id/benchmarks/math/, records of October 2026.