GraySoft
Projects Models Compare Cloud benchmarks FAQ Download guIDE →
Model Intelligence Sheet

ankk98/dspark-qwen3-8b-block7-Q4_K_M-GGUF overview

DSpark Qwen3 8B draft Q4 K M GGUF Q4 K M GGUF draft for DSpark https://huggingface.co/deepseek ai/dspark qwen3 8b block7 speculative decoding with Qwen3 8B htt…

ggufdsparkqwenspeculative-decodingdeepspecbase_model:Qwen/Qwen3-8Bbase_model:quantized:Qwen/Qwen3-8Blicense:mitendpoints_compatibleregion:usconversational

Runs locally from ~1.42 GB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
0
Likes
0
Pipeline
Author

Repository Files & Downloads

1 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
dspark_qwen3_8b_block7.q4_k_m.ggufGGUFGGUF1.42 GBDownload

Model Details

Model IDankk98/dspark-qwen3-8b-block7-Q4_K_M-GGUF
Authorankk98
Pipeline
Licensemit
Base modeldeepseek-ai/dspark_qwen3_8b_block7,Qwen/Qwen3-8B
Last modified2026-06-30T11:44:40.000Z

Model README

---

license: mit

base_model:

  • deepseek-ai/dspark_qwen3_8b_block7
  • Qwen/Qwen3-8B

tags:

  • dspark
  • qwen
  • speculative-decoding
  • deepspec

---

DSpark Qwen3-8B draft (Q4_K_M GGUF)

Q4_K_M GGUF draft for DSpark speculative decoding with Qwen3-8B. Pair it with any Qwen3-8B target GGUF (e.g. Qwen3-8B-Q4_K_M.gguf).

Converted from deepseek-ai/dspark_qwen3_8b_block7 with llama.cpp (LLM_ARCH_DSPARK). This repo contains only the draft weights (~1.5 GB), not the target model.

llama-cli \
  -m /path/to/Qwen3-8B-Q4_K_M.gguf \
  -md ./dspark_qwen3_8b_block7.q4_k_m.gguf \
  --spec-type draft-dspark \
  --spec-draft-n-max 7 \
  -c 512 -ngl 99 -ngld 99 \
  -p "Your prompt" -n 128 --temp 0

Run ankk98/dspark-qwen3-8b-block7-Q4_K_M-GGUF with guIDE

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.

Download guIDE → · Browse 524k+ models · Compare models

Source: Hugging Face · Compare models