Anbeeld/Qwen3-4B-DSpark-GGUF overview
base model: deepseek ai/dspark qwen3 4b block7 tags: safetensors qwen3 region:us Qwen3 4B DSpark GGUF GGUF quantizations of deepseek ai DSpark draft model http…
Runs locally from ~522.6 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| Qwen3-4B-DSpark-Q2_K.gguf | GGUF | Q2_K | 522.6 MB | Download |
| Qwen3-4B-DSpark-Q3_K_M.gguf | GGUF | Q3_K_M | 586.9 MB | Download |
| Qwen3-4B-DSpark-Q4_K_M.gguf | GGUF | Q4_K_M | 653.6 MB | Download |
| Qwen3-4B-DSpark-Q5_K_M.gguf | GGUF | Q5_K_M | 720.4 MB | Download |
| Qwen3-4B-DSpark-Q6_K.gguf | GGUF | Q6_K | 791.3 MB | Download |
| Qwen3-4B-DSpark-Q8_0.gguf | GGUF | Q8_0 | 1023.3 MB | Download |
| Qwen3-4B-DSpark-bf16.gguf | GGUF | BF16 | 1.88 GB | Download |
Model Details
Model README
---
base_model: deepseek-ai/dspark_qwen3_4b_block7
tags:
- safetensors
- qwen3
- region:us
---
Qwen3-4B DSpark GGUF
GGUF quantizations of deepseek-ai DSpark draft model for Qwen3-4B.
Use with BeeLlama.cpp, a llama.cpp fork with advanced quantization features.
---
Qwen3-4B-DSpark
A DSpark speculative-decoding draft model for Qwen/Qwen3-4B, released with the DSpark paper (see the upstream repo deepseek-ai/dspark_qwen3_4b_block7 for checkpoint details). It is not a standalone language model; pair it with the target model in a speculative decoding server.
Run Anbeeld/Qwen3-4B-DSpark-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models