FusionNet 34Bx2 MoE GGUF Benchmarks
HF Avg Score: 67.6 vs 86.1 (GPT-4) -21.5%
ARC: 64.2-33.3%
HellaSwag: 72.4-24%
MMLU: 67.8-21.6%
TruthfulQA: 68.516.1%
WinoGrande: 82.3-5.9%
GSM8K: 51-44.6%
nn.n% — How the model compares to the GPT-4.
Rank the FusionNet 34Bx2 MoE GGUF 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! 🌟
Best Alternatives to FusionNet 34Bx2 MoE GGUF
Best Alternatives |
HF Rank |
Context/RAM |
Downloads |
Likes |
|---|---|---|---|---|
| MixTAO 7Bx2 MoE V8.1 GGUF | 68 | 0K / 4.8 GB | 746 | 10 |
| ...AO 7Bx2 MoE Instruct V7.0 GGUF | 67.1 | 0K / 4.8 GB | 335 | 10 |
| Helion 4x34B GGUF | 66.2 | 0K / 41.5 GB | 32 | 3 |
| Cosmosis 3x34B GGUF | 66.1 | 0K / 31.9 GB | 83 | 6 |
| ...Top 5x7B Instruct S5 V0.1 GGUF | 65.9 | 0K / 2.7 GB | 171 | 1 |
| ...eTop 5x7B Instruct T V0.1 GGUF | 65.7 | 0K / 2.7 GB | 179 | 0 |
| ...Top 5x7B Instruct S4 V0.1 GGUF | 65.7 | 0K / 2.7 GB | 173 | 0 |
| Go Bruins V2.1.1 GGUF | 65.7 | 0K / 3.1 GB | 358 | 7 |
| Quantum DPO V0.1 GGUF | 65.7 | 0K / 3.1 GB | 232 | 1 |
| Quantum V0.01 GGUF | 65.5 | 0K / 3.1 GB | 235 | 2 |
Note: green Score (e.g. "73.2") means that the model is better than TheBloke/FusionNet_34Bx2_MoE-GGUF .
FusionNet 34Bx2 MoE GGUF Parameters and Internals
| LLM Name | FusionNet 34Bx2 MoE GGUF |
| Repository | Open on 🤗 |
| Model Name | FusionNet 34Bx2 MoE |
| Model Creator | Suqin Zhang |
| Base Model(s) | |
| Required VRAM | 22.4 GB |
| Updated | 2024-07-01 |
| Maintainer | TheBloke |
| Model Type | mixtral |
| Model Files | |
| Supported Languages | en |
| GGUF Quantization | Yes |
| Quantization Type | gguf |
| Model Architecture | AutoModel |
| License | mit |
What open-source LLMs or SLMs are you in search of? 34238 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024042801