L3.1 MoE 2x8B V0.2 is an open-source language model by moeru-ai. Features: 8b LLM, VRAM: 27.3GB, Context: 128K, License: llama3.1, MoE, LLM Explorer Score: 0.25.
Base model:arliai/llama-3.1-8b... Base model:joseph717171/llama-... Base model:merge:arliai/llama-... Base model:merge:joseph717171/... Conversational Endpoints compatible Frankenmoe Merge Mergekit Mixtral Model-index Moe Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/moeru-ai/L3.1-Moe-2x8B-v0.2
L3.1 MoE 2x8B V0.2 Benchmarks
MMLU Pro: 31.76
GPQA: 6.71
MUSR: 11.38
BBH: 32.95
IFEval: 73.48 vs 88 (so35)-16.5%
MATH Lvl 5: 16.99
LLME Score: 0.2453
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)
L3.1 MoE 2x8B V0.2 Parameters and Internals
| LLM Name | L3.1 MoE 2x8B V0.2 |
| Repository 🤗 | https://huggingface.co/moeru-ai/L3.1-Moe-2x8B-v0.2 |
| Base Model(s) | |
| Model Size | 8b |
| Required VRAM | 27.3 GB |
| Updated | 2026-04-05 |
| Maintainer | moeru-ai |
| Model Type | mixtral |
| Model Files | |
| Model Architecture | MixtralForCausalLM |
| License | llama3.1 |
| Context Length | 131072 |
| Model Max Length | 131072 |
| Transformers Version | 4.45.2 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Padding Token | <|begin_of_text|> |
| Vocabulary Size | 128256 |
| Torch Data Type | bfloat16 |
Best Alternatives to L3.1 MoE 2x8B V0.2
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...ama 3 Aplite Instruct 4x8B MoE | 8K / 50 GB | 36 | 39 |
| Lamma3merge3 15B MoE | 8K / 27.5 GB | 11 | 1 |
| Lamma3merge2 15B MoE | 8K / 27.5 GB | 10 | 0 |
| Llama3merge7 15B MoE | 8K / 27.5 GB | 7 | 0 |
| Mergkit 1 | 8K / 22.6 GB | 8 | 0 |
| Llama 3 8B Shisa 2x8B | 8K / 7.4 GB | 4 | 2 |
| Llama3merge8 15B MoE | 8K / 27.5 GB | 5 | 0 |
| Llama3merge6 15B MoE | 8K / 27.5 GB | 5 | 0 |
| ...8B Finetune All V6 Epoch2 V0.1 | 2K / 18 GB | 4 | 1 |
| ...oE 8B Pretrain 0520 Iter134999 | 2K / 18 GB | 15 | 0 |
Rank the L3.1 MoE 2x8B V0.2 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