Model Card on HF 🤗: https://huggingface.co/Dans-DiscountModels/12b-mn-dans-reasoning-test-3
12B Mn Dans Reasoning Test 3 Benchmarks
MMLU Pro: 16.84
GPQA: 2.8
MUSR: 11
BBH: 25.85
IFEval: 50.53 vs 88 (so35)-42.6%
MATH Lvl 5: 7.78
LLME Score: 0.27599
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
12B Mn Dans Reasoning Test 3 Parameters and Internals
| LLM Name | 12B Mn Dans Reasoning Test 3 |
| Repository 🤗 | https://huggingface.co/Dans-DiscountModels/12b-mn-dans-reasoning-test-3 |
| Model Size | 12b |
| Required VRAM | 24.5 GB |
| Updated | 2025-07-10 |
| Maintainer | Dans-DiscountModels |
| Model Type | mistral |
| Model Files | |
| Model Architecture | MistralForCausalLM |
| Context Length | 1024000 |
| Model Max Length | 1024000 |
| Transformers Version | 4.49.0 |
| Tokenizer Class | PreTrainedTokenizer |
| Padding Token | <pad> |
| Vocabulary Size | 131072 |
| Torch Data Type | bfloat16 |
Best Alternatives to 12B Mn Dans Reasoning Test 3
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...r Nemo 12B Instruct R 21 09 24 | 1000K / 24.5 GB | 694510 | 126 |
| Captain Eris Violet V0.420 12B | 1000K / 24.5 GB | 4065 | 49 |
| MN 12B Mag Mell R1 | 1000K / 24.5 GB | 5465 | 184 |
| Captain Eris BMO Violent 12B | 1000K / 24.5 GB | 305 | 2 |
| Dans SakuraKaze V1.0.0 12B | 1000K / 24.5 GB | 36 | 18 |
| ...n Eris BMO Violent GRPO V0.420 | 1000K / 24.5 GB | 92 | 3 |
| PLLuM 12B Nc Chat | 1000K / 24.5 GB | 2598 | 7 |
| Mistral Nemo Kartoffel 12B | 1000K / 24.5 GB | 56 | 6 |
| ...s PersonalityEngine V1.1.0 12B | 1000K / 24.5 GB | 80 | 42 |
| DoublePotato Mistral Nemo 13B | 1000K / 26.8 GB | 4 | 1 |
Note: green Score (e.g. "73.2") means that the model is better than Dans-DiscountModels/12b-mn-dans-reasoning-test-3 .
Rank the 12B Mn Dans Reasoning Test 3 Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 50200 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124