OpenHathi 7B Hi v0.1 Base needs about 8.7 GB of memory at Q4_0 (a 3.9 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 2 quants, from 3.9 GB (Q4_0) to 4.3 GB (Q4_1).
OpenHathi 7B Hi v0.1 Base – a Hindi‑English Llama2‑based model
OpenHathi 7B Hi v0.1 Base is a 6.87 billion‑parameter Llama architecture model trained on Hindi, English, and Hinglish. It is a base model and should be fine‑tuned for specific tasks. The model is licensed under llama2.
Good at
- Hindi and Hinglish text generation
- English language tasks
- Multilingual dialogue
Limits
- No vision support
- Requires fine‑tuning for best results
- Context length limited to 4096 tokens
Best for
- Fine‑tuning for Hindi‑English chatbots
- Multilingual content creation
- Research on LLMs in South‑Asian languages
Who it’s for. Researchers and developers looking to build or experiment with a Hindi‑English Llama2‑based foundation model. The model is intended for fine‑tuning rather than direct deployment.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_0 | 3.9 GB | 6.6 GB | Fits | Get file → | |
| Q4_1 | 4.3 GB | 7.0 GB | Fits | Get file → |
For 8 GB VRAM, use the Q4_0 quantization (file size 3.92 GB, estimated memory 8.7 GB). If 9 GB VRAM is available, Q4_1 (4.34 GB, 9.1 GB memory) can be used. Both are approximate estimates.
What GPU do you need for OpenHathi 7B Hi v0.1 Base?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is Q4_1 (4.3 GB file, about 9.1 GB in use).
Which OpenHathi 7B Hi v0.1 Base quant should you choose?
Q4_0 (3.9 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about OpenHathi 7B Hi v0.1 Base
How much VRAM does OpenHathi 7B Hi v0.1 Base need?
About 8.7 GB at Q4_0 with an 8K context window. That is the 3.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run OpenHathi 7B Hi v0.1 Base on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 8.7 GB, so you would need a larger GPU or CPU offloading.
What is the file size of OpenHathi 7B Hi v0.1 Base GGUF?
It depends on the quant: Q4_0 is 3.9 GB, Q4_1 is 4.3 GB.
What context length does OpenHathi 7B Hi v0.1 Base support?
The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 4.3 GB at 8K and 2.1 GB at 4K.
Where can I download OpenHathi 7B Hi v0.1 Base GGUF?
The files are on Hugging Face in the sarvamai/OpenHathi-7B-Hi-v0.1-Base repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does OpenHathi 7B Hi v0.1 Base use?
Hugging Face lists the license as llama2. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run this model on 8 GB VRAM?
Yes, the Q4_0 quantization requires about 8.7 GB of VRAM, which fits within an 8 GB GPU with some headroom.
Does this model support vision tasks?
No, the model has no vision capabilities and is purely text‑based.
Do I need to fine‑tune before using it?
Yes, the base model is not intended for direct use; fine‑tuning on your target tasks is recommended.