Llama 3 8B Claudstruct V3 is an open-source language model by jrahn. Features: 8b LLM, VRAM: 0.1GB, License: llama3, Instruction-Based, HF Score: 65.6, LLM Explorer Score: 0.15, Arc: 59, HellaSwag: 80.1, MMLU: 64.6, TruthfulQA: 51.8, WinoGrande: 74.2, GSM8K: 64.2.
4-bit Adapter Base model:adapter:meta-llama/... Base model:meta-llama/meta-lla... Bitsandbytes Dataset:norquinal/claude multi... En Finetuned Generated from trainer Instruct Llama Lora Peft Region:us Safetensors
Model Card on HF 🤗: https://huggingface.co/jrahn/llama-3-8b-claudstruct-v3
Llama 3 8B Claudstruct V3 Benchmarks
ARC: 58.96 vs 96.7 (so35)-39%
HellaSwag: 80.05 vs 95.3 (gpt4)-16%
MMLU: 64.55 vs 88.3 (so35)-26.9%
TruthfulQA: 51.76 vs 59 (gpt4)-12.3%
WinoGrande: 74.19 vs 87.5 (gpt4)-15.2%
GSM8K: 64.22 vs 96.4 (so35)-33.4%
LLME Score: 0.15016
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)
Llama 3 8B Claudstruct V3 Parameters and Internals
| Model Type |
| |||||||||
| Additional Notes |
| |||||||||
| Training Details |
|
| LLM Name | Llama 3 8B Claudstruct V3 |
| Repository 🤗 | https://huggingface.co/jrahn/llama-3-8b-claudstruct-v3 |
| Base Model(s) | |
| Model Size | 8b |
| Required VRAM | 0.1 GB |
| Updated | 2025-09-23 |
| Maintainer | jrahn |
| Instruction-Based | Yes |
| Model Files | |
| Supported Languages | en |
| Model Architecture | Adapter |
| License | llama3 |
| Is Biased | none |
| Tokenizer Class | PreTrainedTokenizerFast |
| Padding Token | <|end_of_text|> |
| PEFT Type | LORA |
| LoRA Model | Yes |
| PEFT Target Modules | v_proj|up_proj|gate_proj|down_proj|o_proj|k_proj|q_proj |
| LoRA Alpha | 16 |
| LoRA Dropout | 0.05 |
| R Param | 8 |
Best Alternatives to Llama 3 8B Claudstruct V3
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ... 3 8B Instruct Bvr Finetune V3 | 8K / 16.1 GB | 7 | 0 |
| Flippa V6 | 0K / 0 GB | 83 | 1 |
| ...B Instruct DPO 0R100L PoliTune | 0K / 16.1 GB | 6 | 0 |
| ...B Lora Rag Citation Generation | 0K / 0 GB | 10 | 4 |
| ...a 8B Rank 128 INSTRUCT Adapter | 0K / 0.7 GB | 18 | 1 |
| ...lama 3 1 8B Instruct Orca ORPO | 0K / 0.1 GB | 8 | 2 |
| Vortex2 | 0K / 4.4 GB | 8 | 0 |
| Llama3 8B Instruct Code | 0K / 0.2 GB | 4 | 1 |
| ...Llama 3 8B Instruct ORPO QLoRA | 0K / 0.7 GB | 56 | 0 |
| Llama 3 8B Claudstruct V1 | 0K / 0.1 GB | 5 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than jrahn/llama-3-8b-claudstruct-v3 .
Rank the Llama 3 8B Claudstruct V3 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? 53185 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