Dream is a masked diffusion language model based on the Qwen2.5-7B backbone with bidirectional attention and Grouped Query Attention (GQA, 28 query heads / 4 KV heads).
Available Quantizations
File
Type
Size
Description
dream-7b-f16.gguf
F16
~15 GB
Full precision, best quality
dream-7b-q8_0.gguf
Q8_0
~8.2 GB
8-bit quantization, near-lossless
dream-7b-q4km.gguf
Q4_K_M
~5.0 GB
4-bit mixed quantization, best quality/size ratio
Recommended:
Q4_K_M for most users. Q8_0 if you have enough RAM and want minimal quality loss.
Performance
Benchmarked on diffuse-cpp with entropy_exit + inter-step KV cache, 12 threads, seed=42:
Prompt
tok/s
Steps
vs llama.cpp
Capital of France?
21.6
2
2.5x
Translate to French
14.3
6
1.7x
15 x 23?
21.6
2
2.5x
Translate to Spanish
13.2
10
1.6x
Python is_prime()
8.2
7
1.0x
Why sky blue?
4.9
16
0.6x
List planets
4.9
16
0.6x
Poem about ocean
4.5
16
0.5x
Average
11.6
1.4x
Easy prompts (factual, math):
14-22 tok/s
(1.6-2.5x faster than llama.cpp)
Hard prompts (creative, long-form):
4.5-4.9 tok/s
llama.cpp baseline: 8.51 tok/s (Qwen2.5-7B-Instruct, Q4_K_M, same hardware)
Usage
# Download
huggingface-cli download diffuse-cpp/Dream-v0-Instruct-7B-GGUF dream-7b-q4km.gguf
# Run (requires diffuse-cpp v0.2.0+)
./diffuse-cli -m dream-7b-q4km.gguf -p "What is the capital of France?" -n 64 -s 16
Model Details
Architecture:
Qwen2.5-7B backbone with bidirectional attention
Parameters:
7.62B
Layers:
28
Hidden size:
3584
Attention:
GQA (28 query heads, 4 KV heads, head dim 128)
FFN:
SwiGLU, intermediate size 18944
Vocabulary:
152,064 tokens
RoPE theta:
1,000,000
Mask token ID:
151666
Training:
Masked diffusion on text, with autoregressive logit shift
Conversion Details
Converted from SafeTensors using
convert-dream.py
from diffuse-cpp:
339 tensors total (255 weights + 84 QKV biases)
QKV biases kept at F32 in all quantizations
Edge layers (first/last) quantized to Q6_K in Q4_K_M scheme
Citation
@misc{dream2025,
title={Dream 7B - Scalable Discrete Denoising Diffusion Models for Text Generation},
author={Ye, Jiacheng and others},
year={2025}
}
License
Apache 2.0, following the original Dream model license.
Runs of diffuse-cpp Dream-v0-Instruct-7B-GGUF on huggingface.co
102
Total runs
-4
24-hour runs
-2
3-day runs
21
7-day runs
-118
30-day runs
More Information About Dream-v0-Instruct-7B-GGUF huggingface.co Model
More Dream-v0-Instruct-7B-GGUF license Visit here:
Dream-v0-Instruct-7B-GGUF huggingface.co is an AI model on huggingface.co that provides Dream-v0-Instruct-7B-GGUF's model effect (), which can be used instantly with this diffuse-cpp Dream-v0-Instruct-7B-GGUF model. huggingface.co supports a free trial of the Dream-v0-Instruct-7B-GGUF model, and also provides paid use of the Dream-v0-Instruct-7B-GGUF. Support call Dream-v0-Instruct-7B-GGUF model through api, including Node.js, Python, http.
Dream-v0-Instruct-7B-GGUF huggingface.co is an online trial and call api platform, which integrates Dream-v0-Instruct-7B-GGUF's modeling effects, including api services, and provides a free online trial of Dream-v0-Instruct-7B-GGUF, you can try Dream-v0-Instruct-7B-GGUF online for free by clicking the link below.
diffuse-cpp Dream-v0-Instruct-7B-GGUF online free url in huggingface.co:
Dream-v0-Instruct-7B-GGUF is an open source model from GitHub that offers a free installation service, and any user can find Dream-v0-Instruct-7B-GGUF on GitHub to install. At the same time, huggingface.co provides the effect of Dream-v0-Instruct-7B-GGUF install, users can directly use Dream-v0-Instruct-7B-GGUF installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
Dream-v0-Instruct-7B-GGUF install url in huggingface.co:
