Rose 20B 3bpw EXL2 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)
Rose 20B 3bpw EXL2 Parameters and Internals
| LLM Name | Rose 20B 3bpw EXL2 |
| Repository 🤗 | https://huggingface.co/Kooten/Rose-20B-3bpw-exl2 |
| Model Size | 20b |
| Required VRAM | 7.9 GB |
| Updated | 2024-09-07 |
| Maintainer | Kooten |
| Model Type | llama |
| Model Files | |
| Quantization Type | exl2 |
| Model Architecture | LlamaForCausalLM |
| License | cc-by-nc-4.0 |
| Context Length | 4096 |
| Model Max Length | 4096 |
| Transformers Version | 4.36.0.dev0 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | [PAD] |
| Vocabulary Size | 32000 |
| Torch Data Type | float16 |
Best Alternatives to Rose 20B 3bpw EXL2
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| InternLM2 Chat 20B ToxicRP | 32K / 39.9 GB | 5 | 0 |
| ...a Quality Float32 4 25bpw EXL2 | 4K / 10.9 GB | 4 | 2 |
| ...ra Quality Float32 8 0bpw EXL2 | 4K / 17.5 GB | 4 | 1 |
| ...ra Quality Float32 5 0bpw EXL2 | 4K / 12.8 GB | 6 | 0 |
| ... Unquirked 20B 4.25bpw H8 EXL2 | 4K / 11 GB | 4 | 1 |
| ...a No Robots 20B 3.0bpw H6 EXL2 | 4K / 7.8 GB | 3 | 1 |
| ...Cetacean 20B V2.4.0bpw H6 EXL2 | 4K / 10.3 GB | 4 | 1 |
| DaringMaid 20B V1.1 4bpw EXL2 | 4K / 10.4 GB | 5 | 2 |
| DaringMaid 20B V1.1 3bpw EXL2 | 4K / 7.9 GB | 4 | 1 |
| DaringMaid 20B 6bpw EXL2 | 4K / 15.3 GB | 8 | 2 |
Note: green Score (e.g. "73.2") means that the model is better than Kooten/Rose-20B-3bpw-exl2 .
Rank the Rose 20B 3bpw EXL2 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? 35693 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072803