Hala 9B needs about 8.7 GB of memory at Q4_K_M (a 5.4 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 25 quants, from 0.0 GB to 7.2 GB (Q6_K).
Hala 9B – 8.78B‑parameter Gemma2 model for Arabic instruction and translation
Hala 9B is a 8.78B‑parameter Gemma2 model trained on Arabic‑centric instruction and translation data. It supports reasoning tasks and is released under a cc‑by‑nc‑4.0 license. The model is available in multiple GGUF quantizations.
Good at
- Arabic instruction and translation
- Reasoning tasks
- Low‑latency inference with quantized models
Limits
- No vision support
- License restricts commercial use
- Context length limited to 4096 tokens
Best for
- Local Arabic chatbot deployment
- Arabic text summarization
- Instruction‑following in Arabic
Who it’s for. Developers building Arabic‑centric LLM applications who need a reasonably sized model with reasoning capabilities. The model is suitable for research and non‑commercial use.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.0 GB | 1.9 GB | Fits | Get file → | |
| IQ1_S | 2.1 GB | 4.0 GB | Fits | Get file → | |
| IQ1_M | 2.2 GB | 4.1 GB | Fits | Get file → | |
| IQ2_XXS | 2.5 GB | 4.4 GB | Fits | Get file → | |
| IQ2_XS | 2.7 GB | 4.7 GB | Fits | Get file → | |
| IQ2_S | 2.9 GB | 4.8 GB | Fits | Get file → | |
| IQ2_M | 3.1 GB | 5.0 GB | Fits | Get file → | |
| Q2_K_S | 3.2 GB | 5.1 GB | Fits | Get file → | |
| Q2_K | 3.4 GB | 5.3 GB | Fits | Get file → | |
| IQ3_XXS | 3.5 GB | 5.4 GB | Fits | Get file → | |
| IQ3_XS | 3.8 GB | 5.7 GB | Fits | Get file → | |
| IQ3_S | 4.0 GB | 5.9 GB | Fits | Get file → | |
| Q3_K_S | 4.0 GB | 5.9 GB | Fits | Get file → | |
| IQ3_M | 4.1 GB | 6.0 GB | Fits | Get file → | |
| Q3_K_M | 4.4 GB | 6.3 GB | Fits | Get file → | |
| Q3_K_L | 4.8 GB | 6.7 GB | Fits | Get file → | |
| IQ4_XS | 4.8 GB | 6.7 GB | Fits | Get file → | |
| IQ4_NL | 5.1 GB | 7.0 GB | Fits | Get file → | |
| Q4_0 | 5.1 GB | 7.0 GB | Fits | Get file → | |
| Q4_K_S | 5.1 GB | 7.0 GB | Fits | Get file → | |
| Q4_K_M | 5.4 GB | 7.3 GB | Tight | Get file → | |
| Q4_1 | 5.6 GB | 7.5 GB | Tight | Get file → | |
| Q5_K_S | 6.1 GB | 8.0 GB | Too big | Get file → | |
| Q5_K_M | 6.3 GB | 8.2 GB | Too big | Get file → | |
| Q6_K | 7.2 GB | 9.1 GB | Too big | Get file → |
For 8 GB VRAM, the Q4_K_M quant (5.38 GB file, 8.7 GB memory at 8k context) is a good balance. If 16 GB VRAM is available, Q5_K_M (6.27 GB file, 9.6 GB memory) offers higher quality. The smallest IQ1_S (2.06 GB file, 5.4 GB memory) fits 4 GB VRAM setups.
What GPU do you need for Hala 9B?
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: the best quant that fits is the main file (0.0 GB file, about 3.3 GB in use).
- 6 GB VRAM: the best quant that fits is IQ2_XXS (2.5 GB file, about 5.8 GB in use).
- 8 GB VRAM: the best quant that fits is Q3_K_M (4.4 GB file, about 7.7 GB in use).
- 12 GB VRAM: the best quant that fits is Q6_K (7.2 GB file, about 10.5 GB in use).
Which Hala 9B quant should you choose?
Q4_K_M is the usual starting point for Hala 9B: it balances file size and quality well. Here it is a 5.4 GB file.
If memory is tight, the smallest file is the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Hala 9B
How much VRAM does Hala 9B need?
About 8.7 GB at Q4_K_M with an 8K context window. That is the 5.4 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 Hala 9B on an 8 GB GPU?
Yes, with Q3_K_M: the 4.4 GB file needs about 7.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Hala 9B GGUF?
It depends on the quant: file is 0.0 GB, IQ1_S is 2.1 GB, IQ1_M is 2.2 GB, IQ2_XXS is 2.5 GB, IQ2_XS is 2.7 GB, IQ2_S is 2.9 GB and 19 more.
What context length does Hala 9B support?
The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 2.8 GB at 8K and 1.4 GB at 4K.
Where can I download Hala 9B GGUF?
The files are on Hugging Face in the mradermacher/Hala-9B-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Hala 9B use?
Hugging Face lists the license as cc-by-nc-4.0. 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 Hala 9B on 8 GB VRAM?
Yes, the Q4_K_M quant requires about 8.7 GB memory at 8k context, which fits an 8 GB VRAM GPU with some headroom.
Is the model suitable for commercial use?
No, the cc‑by‑nc‑4.0 license prohibits commercial use without permission.
Does Hala 9B support vision tasks?
No, the model does not include vision capabilities.