arunmcops/AyurParam-GGUF

AyurParam — GGUF

Ayurveda · Traditional Medicine  |  BharatGenAI  |  हिन्दी + English

AyurParam is BharatGenAI's Ayurveda and traditional medicine specialist, fine-tuned on classical Ayurvedic texts, herbs, formulations (rasayanas), Panchakarma procedures, and integrative health advisory rooted in Indian medical traditions.

---

Available Quantizations

Quantization	File	Size	Description
F16	AyurParam-F16.gguf	~5.7 GB	Full float 16, near-lossless, GPU recommended
Q2_K	AyurParam-Q2_K.gguf	~0.9 GB	2-bit, smallest, significant quality loss
Q3_K_M	AyurParam-Q3_K_M.gguf	~1.3 GB	3-bit medium, moderate quality
Q4_K_M	AyurParam-Q4_K_M.gguf	~1.7 GB	4-bit medium — recommended for most users
Q5_K_M	AyurParam-Q5_K_M.gguf	~2.0 GB	5-bit medium, great quality
Q6_K	AyurParam-Q6_K.gguf	~2.4 GB	6-bit, excellent quality
Q8_0	AyurParam-Q8_0.gguf	~3.0 GB	8-bit, best quality/size balance

Not sure which to pick? → Q4_K_M for everyday use on CPU/GPU → Q8_0 for best quality on GPU with ≥ 6 GB VRAM → F16/BF16 for full precision research use

---

Use Cases

• Herb and medicinal plant identification
• Dosha analysis and Ayurvedic constitution (Prakriti)
• Classical formulations and Rasayana advisory
• Panchakarma procedure information
• Integrative health and lifestyle guidance

---

Quick Start

llama.cpp

# Windows
winget install llama.cpp

# Run directly from Hugging Face
llama-cli -hf arunmcops/AyurParam-GGUF:AyurParam-Q4_K_M.gguf

# OpenAI-compatible server
llama-server -hf arunmcops/AyurParam-GGUF:AyurParam-Q4_K_M.gguf --port 8080

Ollama

ollama run hf.co/arunmcops/AyurParam-GGUF:AyurParam-Q4_K_M.gguf

Python (llama-cpp-python)

from llama_cpp import Llama

llm = Llama.from_pretrained(
    repo_id="arunmcops/AyurParam-GGUF",
    filename="AyurParam-Q4_K_M.gguf",
)
response = llm.create_chat_completion(
    messages=[{"role": "user", "content": "नमस्ते! आप कौन हैं?"}]
)
print(response["choices"][0]["message"]["content"])

---

About BharatGenAI

BharatGenAI builds open-source language models for India — bilingual (Hindi + English) and domain-specialized for agriculture, healthcare, legal, and general-purpose use.

• HuggingFace: arunmcops

---

License

Apache 2.0 — free for personal, research, and commercial use.