MixtureofMerges MoE 4x7b V5 is an open-source language model by jsfs11. Features: 24.2b LLM, VRAM: 48.3GB, Context: 32K, License: apache-2.0, MoE, HF Score: 76, LLM Explorer Score: 0.22, Arc: 73.9, HellaSwag: 89, MMLU: 64.7, TruthfulQA: 73.7, WinoGrande: 85.1, GSM8K: 69.8.
Base model:eren23/dpo-binarize... Base model:kukedlc/neural4gsm8... Base model:merge:eren23/dpo-bi... Base model:merge:kukedlc/neura... Base model:merge:mlabonne/alph... Base model:merge:paulml/omnibe... Base model:mlabonne/alphamonar... Base model:paulml/omnibeaglesq... Endpoints compatible Eren23/dpo-binarized-neutrixom... Frankenmoe Kukedlc/neural4gsm8k Lazymergekit Merge Mergekit Mixtral Mlabonne/alphamonarch-7b Model-index Moe Paulml/omnibeaglesquaredmbx-v3... Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/jsfs11/MixtureofMerges-MoE-4x7b-v5
MixtureofMerges MoE 4x7b V5 Benchmarks
MMLU Pro: 23.31
GPQA: 4.59
MUSR: 12.34
BBH: 32.83
IFEval: 41.99 vs 88 (so35)-52.3%
ARC: 73.89 vs 96.7 (so35)-23.6%
HellaSwag: 89 vs 95.3 (gpt4)-6.6%
MMLU: 64.69 vs 88.3 (so35)-26.7%
TruthfulQA: 73.73 vs 59 (gpt4)25%
WinoGrande: 85.08 vs 87.5 (gpt4)-2.8%
GSM8K: 69.75 vs 96.4 (so35)-27.6%
MATH Lvl 5: 7.55
LLME Score: 0.21666
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)
MixtureofMerges MoE 4x7b V5 Parameters and Internals
| Model Type |
| |
| Additional Notes |
|
| LLM Name | MixtureofMerges MoE 4x7b V5 |
| Repository 🤗 | https://huggingface.co/jsfs11/MixtureofMerges-MoE-4x7b-v5 |
| Base Model(s) | |
| Model Size | 24.2b |
| Required VRAM | 48.3 GB |
| Updated | 2026-04-22 |
| Maintainer | jsfs11 |
| Model Type | mixtral |
| Model Files | |
| Model Architecture | MixtralForCausalLM |
| License | apache-2.0 |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.38.1 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <s> |
| Vocabulary Size | 32000 |
| Torch Data Type | bfloat16 |
Best Alternatives to MixtureofMerges MoE 4x7b V5
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Dzakwan MoE 4x7b Beta | 32K / 48.4 GB | 9072 | 0 |
| Beyonder 4x7B V3 | 32K / 48.3 GB | 7856 | 60 |
| Calme 4x7B MoE V0.2 | 32K / 48.3 GB | 8019 | 2 |
| Calme 4x7B MoE V0.1 | 32K / 48.3 GB | 7905 | 2 |
| Mera Mix 4x7B | 32K / 48.3 GB | 8157 | 19 |
| MixtureofMerges MoE 4x7b V4 | 32K / 48.3 GB | 7889 | 4 |
| CognitiveFusion2 4x7B BF16 | 32K / 48.3 GB | 8076 | 3 |
| Proto Athena 4x7B | 32K / 48.4 GB | 5 | 0 |
| Proto Athena V0.2 4x7B | 32K / 48.4 GB | 8 | 0 |
| NeuralStar FusionWriter 4x7b | 32K / 48.3 GB | 24 | 9 |
Note: green Score (e.g. "73.2") means that the model is better than jsfs11/MixtureofMerges-MoE-4x7b-v5 .
Rank the MixtureofMerges MoE 4x7b V5 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? 53254 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