Arxiv:1910.09700 4-bit Autotrain compatible Bitsandbytes Conversational Endpoints compatible Instruct Lora Mistral Region:us Safetensors
Model Card on HF 🤗: https://huggingface.co/Trubnik1967/zephyr-7b-beta-Agent-Instruct_v3
Zephyr 7B Beta Agent Instruct V3 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)
Zephyr 7B Beta Agent Instruct V3 Parameters and Internals
| LLM Name | Zephyr 7B Beta Agent Instruct V3 |
| Repository 🤗 | https://huggingface.co/Trubnik1967/zephyr-7b-beta-Agent-Instruct_v3 |
| Model Size | 7b |
| Required VRAM | 0.7 GB |
| Updated | 2025-06-09 |
| Maintainer | Trubnik1967 |
| Instruction-Based | Yes |
| Model Files | |
| Model Architecture | AutoModelForCausalLM |
| Is Biased | none |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | </s> |
| PEFT Type | LORA |
| LoRA Model | Yes |
| PEFT Target Modules | o_proj|down_proj|v_proj|gate_proj|up_proj|k_proj|q_proj |
| LoRA Alpha | 16 |
| LoRA Dropout | 0.1 |
| R Param | 64 |
Best Alternatives to Zephyr 7B Beta Agent Instruct V3
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Mistral 7B Instruct V0.3 | 32K / 14.5 GB | 998656 | 1779 |
| Mistral 7B Instruct V0.3 | 32K / 14.5 GB | 1293 | 5 |
| Mistral 7B Instruct V0.3 | 32K / 14.5 GB | 1116 | 3 |
| ...ralai Mistral 7B Instruct V0.3 | 32K / 14.5 GB | 59 | 2 |
| Mistral 7B Instruct V0.3 | 32K / 14.5 GB | 12 | 0 |
| Full V4 Astromistral Final | 32K / 4.5 GB | 17 | 1 |
| Monarch 1 | 0K / 14.5 GB | 13 | 1 |
| Mistral 7B Instruct Uz | 0K / 14.5 GB | 128 | 14 |
| Cuscuz 7B | 0K / 14.4 GB | 14 | 1 |
| Code Llama Python | 0K / 0.3 GB | 4 | 1 |
Note: green Score (e.g. "73.2") means that the model is better than Trubnik1967/zephyr-7b-beta-Agent-Instruct_v3 .
Rank the Zephyr 7B Beta Agent Instruct 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? 48046 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124