MiniCPM MoE 8x2B Benchmarks
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)
MiniCPM MoE 8x2B Parameters and Internals
| LLM Name | MiniCPM MoE 8x2B |
| Repository 🤗 | https://huggingface.co/openbmb/MiniCPM-MoE-8x2B |
| Model Size | 2b |
| Required VRAM | 27.7 GB |
| Updated | 2024-10-18 |
| Maintainer | openbmb |
| Model Files | |
| Model Architecture | MiniCPMForCausalLM |
| Context Length | 4096 |
| Model Max Length | 4096 |
| Transformers Version | 4.36.0 |
| Tokenizer Class | LlamaTokenizer |
| Vocabulary Size | 122753 |
| Torch Data Type | bfloat16 |
Best Alternatives to MiniCPM MoE 8x2B
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| MiniCPM 2B 128K | 64K / 6 GB | 2327 | 41 |
| MiniCPM 2B Sft Fp32 | 4K / 10.9 GB | 1681 | 295 |
| MiniCPM 2B Sft Bf16 | 4K / 5.5 GB | 1757 | 116 |
| ...iCPM 2B RAFT Lora Hotpotqa Dev | 4K / 5.5 GB | 10 | 0 |
| MiniCPM Duplex | 4K / 5.5 GB | 5 | 2 |
| MiniCPM 2B DPO Bf16 | 4K / 5.5 GB | 1074 | 44 |
| ...iniCPM 2B DPO Fp32 Safetensors | 4K / 10.9 GB | 19 | 1 |
| ...iniCPM 2B DPO Bf16 Safetensors | 4K / 5.5 GB | 14 | 1 |
| ...iniCPM 2B Sft Fp32 Safetensors | 4K / 10.9 GB | 7 | 1 |
| ...iniCPM 2B Sft Fp32 Safetensors | 4K / 10.9 GB | 6 | 1 |
Note: green Score (e.g. "73.2") means that the model is better than openbmb/MiniCPM-MoE-8x2B .
Rank the MiniCPM MoE 8x2B 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? 36966 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072803