Qwen3 Coder Next Dflash, self-quantized to GGUF by Atomic Chat. Built straight from Z Lab's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline.
Highlights
- 0.5B parameters: the weights this repo quantizes.
- Context length: 262,144 tokens (256K), as published by Z Lab.
- 8 layers: Dense decoder.
- Full imatrix ladder: every quant is calibrated with an importance matrix.
[!NOTE] These GGUFs are self-quantized from the original weights, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model.
[!IMPORTANT] Always pass
--jinjaso the Qwen3 Coder Next Dflash chat template is applied. Without it the model can emit malformed turns.
Model Overview
| Property | Value |
|---|---|
| Base model | z-lab/Qwen3-Coder-Next-DFlash |
| Parameters | 0.5B |
| Layers | 8 |
| Context length | 262,144 tokens (256K) |
| Vocabulary | 151,936 |
| Modalities | Text |
| Architecture | Dense decoder, 32 attention heads over 4 KV heads, DFlashDraftModel |
| This repo | GGUF quants (imatrix). Quants: Q8_0 |
Choosing a quant
| Quant | Size | Notes |
|---|---|---|
Q8_0 |
0.5 GB | Effectively lossless, reference quality. |
[!TIP] Pick the largest file that fits your (V)RAM with room for context.
Q8_0is the sweet spot for most setups;Q6_KorQ8_0for maximum fidelity.
Get started
Run Qwen3 Coder Next Dflash locally with:
- Atomic Chat: the easiest path. Open the app, search
AtomicChat/Qwen3-Coder-Next-DFlash-GGUF, pick a quant, hit Use this model. - llama.cpp:
llama-server -hf AtomicChat/Qwen3-Coder-Next-DFlash-GGUF:Q8_0 --jinja -c 8192 - Ollama:
ollama run hf.co/AtomicChat/Qwen3-Coder-Next-DFlash-GGUF:Q8_0 - LM Studio / Jan: search the repo id, download any quant.
Best practices
| Parameter | Value |
|---|---|
| temperature | 0.0 |
Z Lab's recommended sampling configuration for z-lab/Qwen3-Coder-Next-DFlash.
Run in llama.cpp
git clone https://github.com/ggml-org/llama.cpp
cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON
cmake --build llama.cpp/build --config Release -j --target llama-cli llama-server
./llama.cpp/build/bin/llama-server \
-hf AtomicChat/Qwen3-Coder-Next-DFlash-GGUF:Q8_0 \
--jinja -ngl 99 -c 8192 -fa on
How these were made
- Download
z-lab/Qwen3-Coder-Next-DFlash(original weights). - Convert to f16 GGUF with llama.cpp.
- Build an importance matrix over our calibration corpus.
- Quantize the ladder with
--imatrix.
License
Original model by Z Lab, released under the MIT license. Quantized by Atomic Chat.