Gemma 4 31B It DFlash needs about 1.6 GB of memory at Q4_K_M (a 0.9 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 7 quants, from 0.6 GB (Q2_K) to 3.1 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 0.7 GB.
Gemma 4 31B IT DFlash – a lightweight block diffusion drafter
Gemma 4 31B IT DFlash is a draft model that uses a lightweight block diffusion approach to generate multiple tokens in parallel. It is designed to be paired with google/gemma-4-31B-it and is available in several GGUF quantizations. The model supports a trained context of 262144 tokens and is released under the apache-2.0 license.
Good at
- Parallel token drafting
- Speculative decoding
- High throughput with DFlash
Limits
- No vision support
- Requires pairing with gemma-4-31B-it
- Limited to 262144 context tokens
Best for
- Speed‑critical inference
- Batch token generation
- Low‑latency applications
Who it’s for. Developers and researchers who need faster token generation and are willing to pair the drafter with the base Gemma 4 31B IT model. The model is suitable for environments where throughput is prioritized over absolute accuracy.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 0.6 GB | 1.2 GB | Fits | Get file → | |
| Q3_K_M | 0.7 GB | 1.4 GB | Fits | Get file → | |
| Q4_K_M | 0.9 GB | 1.6 GB | Fits | Get file → | |
| Q5_K_M | 1.1 GB | 1.8 GB | Fits | Get file → | |
| Q6_K | 1.3 GB | 1.9 GB | Fits | Get file → | |
| Q8_0 | 1.6 GB | 2.3 GB | Fits | Get file → | |
| BF16 | 3.1 GB | 3.8 GB | Fits | Get file → |
For 8 GB VRAM, Q4_K_M (file size 0.91 GB, memory 1.6 GB) is a good balance. If 4 GB VRAM is available, Q3_K_M (0.75 GB, 1.4 GB) fits. For 2 GB VRAM, Q2_K (0.58 GB, 1.2 GB) is the smallest option. Memory estimates are approximate.
What GPU do you need for Gemma 4 31B It DFlash?
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 BF16 (3.1 GB file, about 3.8 GB in use).
Which Gemma 4 31B It DFlash quant should you choose?
Q4_K_M is the usual starting point for Gemma 4 31B It DFlash: it balances file size and quality well. Here it is a 0.9 GB file.
The Q8_0 file is 1.6 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 Q2_K at 0.6 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 Gemma 4 31B It DFlash
How much VRAM does Gemma 4 31B It DFlash need?
About 1.6 GB at Q4_K_M with an 8K context window. That is the 0.9 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 Gemma 4 31B It DFlash on an 8 GB GPU?
Yes, with BF16: the 3.1 GB file needs about 3.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Gemma 4 31B It DFlash GGUF?
It depends on the quant: Q2_K is 0.6 GB, Q3_K_M is 0.7 GB, Q4_K_M is 0.9 GB, Q5_K_M is 1.1 GB, Q6_K is 1.3 GB, Q8_0 is 1.6 GB and 1 more.
What context length does Gemma 4 31B It DFlash support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.7 GB at 32K.
Where can I download Gemma 4 31B It DFlash GGUF?
The files are on Hugging Face in the Anbeeld/gemma-4-31B-it-DFlash-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 Gemma 4 31B It DFlash 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 Gemma 4 31B IT DFlash on 8 GB VRAM?
Yes, the Q4_K_M quantization uses about 1.6 GB of VRAM at 8 k context, which fits within 8 GB.
Does this model support vision tasks?
No, the model has no vision capabilities.
What license applies to this model?
The model is released under the apache-2.0 license.