CapriAI2 needs about 2.4 GB of memory at Q4_K_M (a 1.6 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 4 quants, from 1.6 GB (Q4_K_M) to 5.3 GB (Q4_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 1.4 GB.
CapriAI2: Fine‑tuned Qwen3.5‑4B with multimodal vision and reasoning
CapriAI2 is a fan‑made fine‑tune of Qwen/Qwen3.5‑4B, optimized for conversational use with vision, reasoning, and tool use. The model is released under Apache‑2.0 and supports multimodal input via a projector file.
Good at
- Conversational reasoning
- Vision‑language tasks
- Tool‑based interactions
Limits
- Requires a projector for image input
- No official language support list
- Large VRAM for full context
Best for
- Chatbots needing vision support
- Rapid prototyping of multimodal agents
- Educational demos of multimodal LLMs
Who it’s for. Developers and hobbyists who want a conversational model with vision capabilities without commercial licensing. It is suitable for research and personal projects where Apache‑2.0 compliance is acceptable.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 1.6 GB | 2.4 GB | Fits | Get file → | |
| Q4_K_M | 2.8 GB | 3.6 GB | Fits | Get file → | |
| Q4_K_M | 5.3 GB | 6.2 GB | Fits | Get file → | |
| Q4_K_M | 5.3 GB | 6.2 GB | Fits | Get file → |
For 8 GB VRAM, use the 1.56 GB Q4_K_M file (≈2.4 GB memory at 8 k context). For 12 GB VRAM, the 2.78 GB file (≈3.6 GB memory) is recommended. If you have 24 GB VRAM, the 5.34 GB file (≈6.2 GB memory) can be used for larger context or higher throughput.
What GPU do you need for CapriAI2?
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 Q4_K_M (2.8 GB file, about 3.6 GB in use).
- 6 GB VRAM: the best quant that fits is Q4_K_M (2.8 GB file, about 3.6 GB in use).
- 8 GB VRAM: the best quant that fits is Q4_K_M (5.3 GB file, about 6.2 GB in use).
Which CapriAI2 quant should you choose?
Q4_K_M is the usual starting point for CapriAI2: it balances file size and quality well. Here it is a 1.6 GB file.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about CapriAI2
How much VRAM does CapriAI2 need?
About 2.4 GB at Q4_K_M with an 8K context window. That is the 1.6 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 CapriAI2 on an 8 GB GPU?
Yes, with Q4_K_M: the 5.3 GB file needs about 6.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of CapriAI2 GGUF?
It depends on the quant: Q4_K_M is 1.6 GB, Q4_K_M is 2.8 GB, Q4_K_M is 5.3 GB, Q4_K_M is 5.3 GB.
What context length does CapriAI2 support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.4 GB at 32K.
Where can I download CapriAI2 GGUF?
The files are on Hugging Face in the Ghasty3/CapriAI2 repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does CapriAI2 use?
Hugging Face lists the license as apache-2.0. 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 CapriAI2 on 8 GB VRAM?
Yes, the 1.56 GB Q4_K_M quant file uses about 2.4 GB memory at 8 k context, fitting within 8 GB VRAM.
Does CapriAI2 support image input?
Image input requires the separate multimodal projector file; the base model alone does not process images.
What license applies to CapriAI2?
CapriAI2 is released under Apache‑2.0, the same license as the base Qwen3.5‑4B model.