Llama Joycaption Beta One Hf Llava needs about 6.5 GB of memory at Q4_K_M (a 4.9 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 24 quants, from 2.0 GB (IQ1_S) to 6.6 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
8B vision model for image captioning
Llama Joycaption Beta One is an 8B parameter vision language model. It generates captions for images and is designed to be free, open, and uncensored. The model supports a 131072 token context window.
Good at
- Image captioning
- Unrestricted content handling
- Wide style coverage
Limits
- No vision support beyond image captioning
- No license restrictions specified
- Large memory requirement for high‑quality quants
Best for
- Captioning photos for diffusion training
- Generating descriptive tags for image datasets
- Automating image metadata creation
Who it’s for. Developers and researchers needing an open, uncensored VLM for captioning tasks. The model is suitable for those building or fine‑tuning diffusion models and requiring broad coverage of image styles and content.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 2.0 GB | 3.6 GB | Fits | Get file → | |
| IQ1_M | 2.2 GB | 3.7 GB | Fits | Get file → | |
| IQ2_XXS | 2.4 GB | 4.0 GB | Fits | Get file → | |
| IQ2_XS | 2.6 GB | 4.2 GB | Fits | Get file → | |
| IQ2_S | 2.8 GB | 4.3 GB | Fits | Get file → | |
| IQ2_M | 2.9 GB | 4.5 GB | Fits | Get file → | |
| Q2_K_S | 3.0 GB | 4.6 GB | Fits | Get file → | |
| Q2_K | 3.2 GB | 4.8 GB | Fits | Get file → | |
| IQ3_XXS | 3.3 GB | 4.8 GB | Fits | Get file → | |
| IQ3_XS | 3.5 GB | 5.1 GB | Fits | Get file → | |
| Q3_K_S | 3.7 GB | 5.2 GB | Fits | Get file → | |
| IQ3_S | 3.7 GB | 5.3 GB | Fits | Get file → | |
| IQ3_M | 3.8 GB | 5.4 GB | Fits | Get file → | |
| Q3_K_M | 4.0 GB | 5.6 GB | Fits | Get file → | |
| Q3_K_L | 4.3 GB | 5.9 GB | Fits | Get file → | |
| IQ4_XS | 4.4 GB | 6.0 GB | Fits | Get file → | |
| Q4_0 | 4.7 GB | 6.2 GB | Fits | Get file → | |
| IQ4_NL | 4.7 GB | 6.3 GB | Fits | Get file → | |
| Q4_K_S | 4.7 GB | 6.3 GB | Fits | Get file → | |
| Q4_K_M | 4.9 GB | 6.5 GB | Fits | Get file → | |
| Q4_1 | 5.1 GB | 6.7 GB | Fits | Get file → | |
| Q5_K_S | 5.6 GB | 7.2 GB | Fits | Get file → | |
| Q5_K_M | 5.7 GB | 7.3 GB | Tight | Get file → | |
| Q6_K | 6.6 GB | 8.2 GB | Too big | Get file → |
For 8‑GB VRAM, use IQ4_XS (6 GB) or Q4_0 (6.2 GB). For 12‑GB VRAM, IQ4_K_M (6.5 GB) or Q5_K_S (7.2 GB) are good choices. For 16‑GB VRAM, Q5_K_M (7.3 GB) or Q6_K (8.2 GB) provide higher quality.
What GPU do you need for Llama Joycaption Beta One Hf Llava?
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 IQ2_XXS (2.4 GB file, about 4.0 GB in use).
- 6 GB VRAM: the best quant that fits is IQ4_XS (4.4 GB file, about 6.0 GB in use).
- 8 GB VRAM: the best quant that fits is Q5_K_M (5.7 GB file, about 7.3 GB in use).
- 12 GB VRAM: the best quant that fits is Q6_K (6.6 GB file, about 8.2 GB in use).
Which Llama Joycaption Beta One Hf Llava quant should you choose?
Q4_K_M is the usual starting point for Llama Joycaption Beta One Hf Llava: it balances file size and quality well. Here it is a 4.9 GB file.
If memory is tight, the smallest file is IQ1_S at 2.0 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about Llama Joycaption Beta One Hf Llava
How much VRAM does Llama Joycaption Beta One Hf Llava need?
About 6.5 GB at Q4_K_M with an 8K context window. That is the 4.9 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 Llama Joycaption Beta One Hf Llava on an 8 GB GPU?
Yes, with Q5_K_M: the 5.7 GB file needs about 7.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Llama Joycaption Beta One Hf Llava GGUF?
It depends on the quant: IQ1_S is 2.0 GB, IQ1_M is 2.2 GB, IQ2_XXS is 2.4 GB, IQ2_XS is 2.6 GB, IQ2_S is 2.8 GB, IQ2_M is 2.9 GB and 18 more.
What context length does Llama Joycaption Beta One Hf Llava support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download Llama Joycaption Beta One Hf Llava GGUF?
The files are on Hugging Face in the mradermacher/llama-joycaption-beta-one-hf-llava-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
Can I run Llama Joycaption on 8 GB VRAM?
Yes, using IQ4_XS or Q4_0 quants, which require about 6 GB of VRAM at 8k context.
Does the model support text‑only inference?
No, it is a vision model and requires image input to function.
Is the model licensed for commercial use?
No license information is provided; use the model at your own risk.