MoE Merging is an open-source language model by Cartinoe5930. Features: 24.2b LLM, VRAM: 48.3GB, Context: 32K, License: apache-2.0, MoE, Instruction-Based, HF Score: 66.8, LLM Explorer Score: 0.14, Arc: 65.4, HellaSwag: 84.6, MMLU: 61.3, TruthfulQA: 57.8, WinoGrande: 77.7, GSM8K: 54.2.
Arxiv:1910.09700 Conversational Endpoints compatible Instruct Mixtral Moe Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/Cartinoe5930/MoE-Merging
MoE Merging Benchmarks
ARC: 65.44 vs 96.7 (so35)-32.3%
HellaSwag: 84.58 vs 95.3 (gpt4)-11.2%
MMLU: 61.31 vs 88.3 (so35)-30.6%
TruthfulQA: 57.83 vs 59 (gpt4)-2%
WinoGrande: 77.66 vs 87.5 (gpt4)-11.2%
GSM8K: 54.21 vs 96.4 (so35)-43.8%
LLME Score: 0.13856
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)
MoE Merging Parameters and Internals
| LLM Name | MoE Merging |
| Repository 🤗 | https://huggingface.co/Cartinoe5930/MoE-Merging |
| Model Size | 24.2b |
| Required VRAM | 48.3 GB |
| Updated | 2026-03-29 |
| Maintainer | Cartinoe5930 |
| Model Type | mixtral |
| Instruction-Based | Yes |
| Model Files | |
| Model Architecture | MixtralForCausalLM |
| License | apache-2.0 |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.37.0 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <s> |
| Vocabulary Size | 32000 |
| Torch Data Type | float16 |
Best Alternatives to MoE Merging
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Proto Athena 4x7B | 32K / 48.4 GB | 5 | 0 |
| Proto Athena V0.2 4x7B | 32K / 48.4 GB | 6 | 0 |
| ...erges MoE 4x7b V10 Mixtralv0.3 | 32K / 48.3 GB | 6 | 0 |
| Sixtyoneeighty 4x7B V1 | 32K / 48.3 GB | 109 | 0 |
| MixtureofMerges MoE 4x7bRP V11 | 32K / 48.3 GB | 13 | 1 |
| ...icon Mixtral87 Merged Torch212 | 32K / 26.7 GB | 6 | 0 |
| Eclipse Mistral 4x7b | 32K / 48.5 GB | 16 | 1 |
| West Ramen 7Bx4 | 32K / 48.7 GB | 85 | 0 |
| Kicon Mixtral87 Merged 41766 | 32K / 26.7 GB | 9 | 0 |
| Boundary Mistral 4x7b MoE | 32K / 48.7 GB | 117 | 1 |
Note: green Score (e.g. "73.2") means that the model is better than Cartinoe5930/MoE-Merging .
Rank the MoE Merging 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? 52721 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a