Autotrain compatible Conversational Custom code En Endpoints compatible Instruct Phi3 Region:us Safetensors Sft Sharded Tensorflow Trl
Model Card on HF 🤗: https://huggingface.co/Giang07/phi3-mini-4k-qlora
Phi3 Mini 4K Qlora 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)
Phi3 Mini 4K Qlora Parameters and Internals
| Model Type |
|
| LLM Name | Phi3 Mini 4K Qlora |
| Repository 🤗 | https://huggingface.co/Giang07/phi3-mini-4k-qlora |
| Model Size | 3.8b |
| Required VRAM | 7.7 GB |
| Updated | 2025-01-15 |
| Maintainer | Giang07 |
| Model Type | phi3 |
| Instruction-Based | Yes |
| Model Files | |
| Supported Languages | en |
| Model Architecture | Phi3ForCausalLM |
| License | mit |
| Context Length | 4096 |
| Model Max Length | 4096 |
| Transformers Version | 4.41.2 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <unk> |
| Vocabulary Size | 32064 |
| Torch Data Type | float16 |
Best Alternatives to Phi3 Mini 4K Qlora
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Phi 3.5 Mini Instruct | 128K / 7.7 GB | 560025 | 752 |
| Phi 3 Mini 128K Instruct | 128K / 7.7 GB | 394305 | 1626 |
| NuExtract V1.5 | 128K / 7.7 GB | 108511 | 89 |
| NuExtract 1.5 | 128K / 7.7 GB | 10379 | 187 |
| Glider | 128K / 15.4 GB | 1575 | 30 |
| Phi 3.5 Mini TitanFusion 0.1 | 128K / 7.7 GB | 18 | 0 |
| ECE EIFFEL 3Bv2 | 128K / 7.7 GB | 13 | 0 |
| Phi3.5 Comets 3.8B | 128K / 7.7 GB | 129 | 0 |
| Samantha2.0 Phi 3.5 Mini ITA | 128K / 7.7 GB | 3950 | 0 |
| Artemide 3.5 | 128K / 7.7 GB | 6817 | 2 |
Note: green Score (e.g. "73.2") means that the model is better than Giang07/phi3-mini-4k-qlora .
Rank the Phi3 Mini 4K Qlora 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? 41418 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227