h2o-danube3-4b-chat needs about 3.6 GB of memory at Q4_K_M (a 2.4 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 16 quants, from 1.4 GB (IQ2_M) to 15.8 GB (F32).
h2o-danube3-4b-chat – a 4 b parameter chat model
h2o-danube3-4b-chat is a chat fine‑tuned model by H2O.ai. It has 3.96 b parameters and a context length of 8 192 tokens. The model is released under the apache‑2.0 license and can be run offline on phones.
Good at
- Chat‑style interactions
- Fine‑tuned responses
- Mistral tokenizer support
Limits
- No vision support
- Requires 8 192 token context
- No system prompt
Best for
- Personal chat assistants
- Lightweight chatbot deployments
- Offline mobile usage
Who it’s for. Developers building chat applications that need a small, offline‑ready model. The model is suitable for mobile or edge devices with limited VRAM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_M | 1.4 GB | 2.7 GB | Fits | Get file → | |
| Q2_K_L | 1.6 GB | 2.9 GB | Fits | Get file → | |
| IQ3_M | 1.8 GB | 3.1 GB | Fits | Get file → | |
| Q3_K_L | 2.1 GB | 3.4 GB | Fits | Get file → | |
| IQ4_XS | 2.2 GB | 3.4 GB | Fits | Get file → | |
| Q3_K_XL | 2.2 GB | 3.5 GB | Fits | Get file → | |
| Q4_K_S | 2.3 GB | 3.5 GB | Fits | Get file → | |
| Q4_K_M | 2.4 GB | 3.6 GB | Fits | Get file → | |
| Q4_K_L | 2.5 GB | 3.7 GB | Fits | Get file → | |
| Q5_K_S | 2.7 GB | 4.0 GB | Fits | Get file → | |
| Q5_K_M | 2.8 GB | 4.1 GB | Fits | Get file → | |
| Q5_K_L | 2.9 GB | 4.1 GB | Fits | Get file → | |
| Q6_K | 3.3 GB | 4.5 GB | Fits | Get file → | |
| Q6_K_L | 3.3 GB | 4.6 GB | Fits | Get file → | |
| Q8_0 | 4.2 GB | 5.5 GB | Fits | Get file → | |
| F32 | 15.8 GB | 17.1 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_K_M (2.39 GB file, ~3.6 GB memory). For 12 GB VRAM, Q5_K_M (2.81 GB file, ~4.1 GB memory). For 16 GB VRAM, Q6_K_L (3.31 GB file, ~4.6 GB memory).
What GPU do you need for h2o-danube3-4b-chat?
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 Q5_K_S (2.7 GB file, about 4.0 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (4.2 GB file, about 5.5 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (4.2 GB file, about 5.5 GB in use).
- 12 GB VRAM: the best quant that fits is Q8_0 (4.2 GB file, about 5.5 GB in use).
- 16 GB VRAM: the best quant that fits is Q8_0 (4.2 GB file, about 5.5 GB in use).
- 24 GB VRAM: the best quant that fits is F32 (15.8 GB file, about 17.1 GB in use).
Which h2o-danube3-4b-chat quant should you choose?
Q4_K_M is the usual starting point for h2o-danube3-4b-chat: it balances file size and quality well. Here it is a 2.4 GB file.
The Q8_0 file is 4.2 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is IQ2_M at 1.4 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
Source
Frequently asked questions about h2o-danube3-4b-chat
How much VRAM does h2o-danube3-4b-chat need?
About 3.6 GB at Q4_K_M with an 8K context window. That is the 2.4 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 h2o-danube3-4b-chat on an 8 GB GPU?
Yes, with Q8_0: the 4.2 GB file needs about 5.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of h2o-danube3-4b-chat GGUF?
It depends on the quant: IQ2_M is 1.4 GB, Q2_K_L is 1.6 GB, IQ3_M is 1.8 GB, Q3_K_L is 2.1 GB, IQ4_XS is 2.2 GB, Q3_K_XL is 2.2 GB and 10 more.
What context length does h2o-danube3-4b-chat support?
The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 0.8 GB at 8K and 0.8 GB at 8K.
Where can I download h2o-danube3-4b-chat GGUF?
The files are on Hugging Face in the bartowski/h2o-danube3-4b-chat-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 h2o-danube3-4b-chat 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 h2o-danube3-4b-chat on 8 GB VRAM?
Yes, use the Q4_K_M quant which requires about 3.6 GB of VRAM at 8 192 context.
Does the model support vision or image input?
No, the model does not support vision or image input.
Is a system prompt supported?
No, the model does not support a system prompt.