GGUF
English
rocmfp4
qwen3next
qwen3-coder-next
coder
Mixture of Experts
imatrix
strix-halo
amd
rocm
vulkan
conversational
Instructions to use plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF # Run inference directly in the terminal: llama cli -hf plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF # Run inference directly in the terminal: llama cli -hf plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF # Run inference directly in the terminal: ./build/bin/llama cli -hf plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
Use Docker
docker model run hf.co/plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
- Llama new
- LM Studio
- Jan
- Ollama
How to use plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF with Ollama:
ollama run hf.co/plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
- Unsloth Desktop
- Pi
How to use plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:9931/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF with Docker Model Runner:
docker model run hf.co/plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
- Lemonade
How to use plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
Run and chat with the model
lemonade run user.Qwen3-Coder-Next-ROCmFP4-GGUF-{{QUANT_TAG}}List all available models
lemonade list
- Hermes Agent
How to use plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:9931/v1 hermes config set model.default plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:9931/v1 \ --custom-model-id "plunderstruck/Qwen3-Coder-Next-ROCmFP4-GGUF" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -85,14 +85,12 @@ Experimental **AMD Strix Halo (gfx1151)** quant of [**Qwen3-Coder-Next**](https:
|
|
| 85 |
<th style="border:1px solid currentColor; padding:7px 10px; text-align:left; text-transform:uppercase; font-size:10px; letter-spacing:1px;">Pick if</th>
|
| 86 |
</tr></thead>
|
| 87 |
<tbody>
|
| 88 |
-
<tr><td style="border:1px solid currentColor; padding:7px 10px;"><code>β¦-
|
| 89 |
-
<tr><td style="border:1px solid currentColor; padding:7px 10px;"><code>β¦-STRIX-embQ8-imatrix-headQ6.gguf</code></td><td style="border:1px solid currentColor; padding:7px 10px;">Q6_K</td><td style="border:1px solid currentColor; padding:7px 10px;">fast body β ~same fidelity, slightly smaller</td></tr>
|
| 90 |
-
<tr><td style="border:1px solid currentColor; padding:7px 10px;"><code>β¦-STRIX-embQ8-imatrix.gguf</code></td><td style="border:1px solid currentColor; padding:7px 10px;">ROCmFP4 4-bit</td><td style="border:1px solid currentColor; padding:7px 10px;">smallest / fastest decode</td></tr>
|
| 91 |
</tbody>
|
| 92 |
</table>
|
| 93 |
</div>
|
| 94 |
|
| 95 |
-
|
| 96 |
|
| 97 |
<div style="border:1px solid currentColor; padding:8px 13px; font-family:ui-monospace,'SF Mono',Consolas,monospace; font-size:12px; margin:12px 0; opacity:0.85;">
|
| 98 |
<b>NOTE //</b> <b>Q8 embeddings</b> (not f16): the source is Q8_0, so Q8 matches its precision exactly β f16 would be fake-f16 bloat for zero gain (embeddings are a lookup, not a matmul).
|
|
@@ -105,7 +103,7 @@ Run from the folder holding the `.gguf` (the Qwen ChatML template is baked in
|
|
| 105 |
```bash
|
| 106 |
env HSA_OVERRIDE_GFX_VERSION=11.5.1 GGML_HIP_ENABLE_UNIFIED_MEMORY=1 \
|
| 107 |
llama-server \
|
| 108 |
-
-m Qwen3-Coder-Next-ROCmFP4-
|
| 109 |
--alias coder-next \
|
| 110 |
--host 0.0.0.0 \
|
| 111 |
--port 8080 \
|
|
@@ -177,30 +175,18 @@ Qwen3-Coder-Next is an **agentic coder** β built to call tools, not narrate co
|
|
| 177 |
<tr><td style="border:1px solid currentColor; padding:8px 11px; width:42%;">DECODE Β· short context</td><td style="border:1px solid currentColor; padding:8px 11px; font-weight:700;">~54 t/s (Vulkan / Ryzen AI Max+ 395)</td></tr>
|
| 178 |
<tr><td style="border:1px solid currentColor; padding:8px 11px;">SPECULATIVE DECODE</td><td style="border:1px solid currentColor; padding:8px 11px; font-weight:700;">none (no MTP head)</td></tr>
|
| 179 |
<tr><td style="border:1px solid currentColor; padding:8px 11px;">LONG CONTEXT</td><td style="border:1px solid currentColor; padding:8px 11px;">cheap β DeltaNet near-constant memory</td></tr>
|
| 180 |
-
<tr><td style="border:1px solid currentColor; padding:8px 11px;">QUANTIZATION</td><td style="border:1px solid currentColor; padding:8px 11px;">
|
| 181 |
</tbody>
|
| 182 |
</table>
|
| 183 |
</div>
|
| 184 |
|
| 185 |
-
**
|
| 186 |
|
| 187 |
-
|
| 188 |
-
<table style="width:100%; border-collapse:collapse; border-radius:0; font-family:ui-monospace,'SF Mono',Consolas,monospace; font-size:12.5px;">
|
| 189 |
-
<thead><tr>
|
| 190 |
-
<th style="border:1px solid currentColor; padding:7px 10px; text-align:left; text-transform:uppercase; font-size:10px; letter-spacing:1px;">Build (imatrix + embQ8 + headQ6)</th>
|
| 191 |
-
<th style="border:1px solid currentColor; padding:7px 10px; text-align:left; text-transform:uppercase; font-size:10px; letter-spacing:1px;">Body</th>
|
| 192 |
-
<th style="border:1px solid currentColor; padding:7px 10px; text-align:left; text-transform:uppercase; font-size:10px; letter-spacing:1px;">Mean KLD vs Q8 β</th>
|
| 193 |
-
<th style="border:1px solid currentColor; padding:7px 10px; text-align:left; text-transform:uppercase; font-size:10px; letter-spacing:1px;">Median KLD β</th>
|
| 194 |
-
<th style="border:1px solid currentColor; padding:7px 10px; text-align:left; text-transform:uppercase; font-size:10px; letter-spacing:1px;">Top-token</th>
|
| 195 |
-
</tr></thead>
|
| 196 |
-
<tbody>
|
| 197 |
-
<tr><td style="border:1px solid currentColor; padding:7px 10px;"><code>COHERENT</code> β
</td><td style="border:1px solid currentColor; padding:7px 10px;">all-dual</td><td style="border:1px solid currentColor; padding:7px 10px;"><b>0.03866</b></td><td style="border:1px solid currentColor; padding:7px 10px;"><b>0.00352</b></td><td style="border:1px solid currentColor; padding:7px 10px;"><b>92.98%</b></td></tr>
|
| 198 |
-
<tr><td style="border:1px solid currentColor; padding:7px 10px;"><code>STRIX</code></td><td style="border:1px solid currentColor; padding:7px 10px;">fast</td><td style="border:1px solid currentColor; padding:7px 10px;">0.04252</td><td style="border:1px solid currentColor; padding:7px 10px;">0.00395</td><td style="border:1px solid currentColor; padding:7px 10px;">92.32%</td></tr>
|
| 199 |
-
</tbody>
|
| 200 |
-
</table>
|
| 201 |
-
</div>
|
| 202 |
|
| 203 |
-
|
|
|
|
|
|
|
| 204 |
|
| 205 |
**Fast even without speculative decoding.** 3B active params + linear Gated-DeltaNet attention β ~54 t/s short-context decode on a Ryzen AI Max+ 395 (Vulkan0), and cheap long context. No MTP needed.
|
| 206 |
|
|
@@ -242,13 +228,10 @@ So the imatrix **measurably improves quantization fidelity to the full model on
|
|
| 242 |
# code-weighted imatrix on the Q8 (single pass; ratio = the real lever)
|
| 243 |
llama-imatrix -m Qwen3-Coder-Next-Q8_0.gguf -f code-weighted-calib.txt -o coder-next.imatrix -c 512 -ngl 999
|
| 244 |
|
| 245 |
-
# quant -> ROCmFP4 with the imatrix (Q8 embeddings)
|
| 246 |
-
|
| 247 |
-
Qwen3-Coder-Next-Q8_0.gguf Qwen3-Coder-Next-ROCmFP4-STRIX-embQ8-imatrix.gguf Q4_0_ROCMFP4_STRIX
|
| 248 |
-
|
| 249 |
-
# RECOMMENDED: the COHERENT all-dual-scale body + Q6 head (the β
file) β lowest KL (Β§04)
|
| 250 |
llama-quantize --allow-requantize --token-embedding-type q8_0 --output-tensor-type q6_K --imatrix coder-next.imatrix \
|
| 251 |
-
Qwen3-Coder-Next-Q8_0.gguf Qwen3-Coder-Next-ROCmFP4-
|
| 252 |
```
|
| 253 |
|
| 254 |
> Experimental research build for AMD Strix Halo β hardware/driver/prompt-sensitive, may not reproduce elsewhere. Not native FP4 tensor-core execution.
|
|
|
|
| 85 |
<th style="border:1px solid currentColor; padding:7px 10px; text-align:left; text-transform:uppercase; font-size:10px; letter-spacing:1px;">Pick if</th>
|
| 86 |
</tr></thead>
|
| 87 |
<tbody>
|
| 88 |
+
<tr><td style="border:1px solid currentColor; padding:7px 10px;"><code>β¦-STRIX-embQ8-imatrix-headQ6.gguf</code> β
</td><td style="border:1px solid currentColor; padding:7px 10px;">Q6_K</td><td style="border:1px solid currentColor; padding:7px 10px;"><b>the one build</b> β best speed/quality balance: Q8 embeddings + Q6 output head on the fast single-scale body</td></tr>
|
|
|
|
|
|
|
| 89 |
</tbody>
|
| 90 |
</table>
|
| 91 |
</div>
|
| 92 |
|
| 93 |
+
One file β the **best speed/quality balance** in ROCmFP4 for Strix Halo. It keeps the two quality levers that are actually *felt* β **Q8 token embeddings** (matching the Q8 source exactly) and a **Q6_K output head** β on the fast single-scale `q4_0_rocmfp4_fast` body + a code-weighted imatrix. Not the most faithful possible (see the fidelity link in Β§04) β it's the point where speed and quality meet best. The DeltaNet-specific tensors (`ssm_conv1d`, `ssm_a`, norms, router) stay **F32**; MoE experts + attention/SSM projections are 4-bit ROCmFP4.
|
| 94 |
|
| 95 |
<div style="border:1px solid currentColor; padding:8px 13px; font-family:ui-monospace,'SF Mono',Consolas,monospace; font-size:12px; margin:12px 0; opacity:0.85;">
|
| 96 |
<b>NOTE //</b> <b>Q8 embeddings</b> (not f16): the source is Q8_0, so Q8 matches its precision exactly β f16 would be fake-f16 bloat for zero gain (embeddings are a lookup, not a matmul).
|
|
|
|
| 103 |
```bash
|
| 104 |
env HSA_OVERRIDE_GFX_VERSION=11.5.1 GGML_HIP_ENABLE_UNIFIED_MEMORY=1 \
|
| 105 |
llama-server \
|
| 106 |
+
-m Qwen3-Coder-Next-ROCmFP4-STRIX-embQ8-imatrix-headQ6.gguf \
|
| 107 |
--alias coder-next \
|
| 108 |
--host 0.0.0.0 \
|
| 109 |
--port 8080 \
|
|
|
|
| 175 |
<tr><td style="border:1px solid currentColor; padding:8px 11px; width:42%;">DECODE Β· short context</td><td style="border:1px solid currentColor; padding:8px 11px; font-weight:700;">~54 t/s (Vulkan / Ryzen AI Max+ 395)</td></tr>
|
| 176 |
<tr><td style="border:1px solid currentColor; padding:8px 11px;">SPECULATIVE DECODE</td><td style="border:1px solid currentColor; padding:8px 11px; font-weight:700;">none (no MTP head)</td></tr>
|
| 177 |
<tr><td style="border:1px solid currentColor; padding:8px 11px;">LONG CONTEXT</td><td style="border:1px solid currentColor; padding:8px 11px;">cheap β DeltaNet near-constant memory</td></tr>
|
| 178 |
+
<tr><td style="border:1px solid currentColor; padding:8px 11px;">QUANTIZATION</td><td style="border:1px solid currentColor; padding:8px 11px;">fast single-scale body + Q8 emb + Q6 head + code-weighted imatrix (measured win β below)</td></tr>
|
| 179 |
</tbody>
|
| 180 |
</table>
|
| 181 |
</div>
|
| 182 |
|
| 183 |
+
**This is the best speed/quality balance in ROCmFP4 β by design, not the absolute fastest.** On top of the imatrix + Q8 emb + Q6 head, we swept the body kernel against the Q8 source by **KL divergence** (the right fidelity metric). An all-dual-scale body did edge the fast single-scale body on KL, but the gain sat inside the measurement noise while costing decode speed β so the **fast single-scale body + Q8 embeddings + Q6 head** is the right point, and the one file we ship.
|
| 184 |
|
| 185 |
+
This mirrors the fuller sweep on our [**Qwen3.6-27B sibling**](https://huggingface.co/plunderstruck/Qwen3.6-27B-MTP-ROCmFP4-GGUF), where every higher-precision body lever (all-dual-scale, selective Q5/Q6 bumps) bought a KL improvement inside the noise at a real speed cost β and where copying an entire dynamic-quant high-precision allocation onto ROCmFP4 *still* couldn't match a true dynamic K-quant, because FP4 is intrinsically less faithful than Q4_K's 4-bit. The same format limit applies here: within ROCmFP4, fast body + Q8 emb + Q6 head is the optimal balance; for maximum fidelity reach for a dynamic K-quant of the base (box below). *(Directional internal measurements β KL vs Q8 on held-out code; reproduce before citing.)*
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 186 |
|
| 187 |
+
<div style="border:1px solid currentColor; padding:8px 13px; font-family:ui-monospace,'SF Mono',Consolas,monospace; font-size:12px; margin:12px 0; opacity:0.9;">
|
| 188 |
+
<b>WANT MAXIMUM FIDELITY INSTEAD OF SPEED?</b> Grab a <b>Q6_K / Q8 dynamic GGUF of the base</b> from <a href="https://huggingface.co/Qwen/Qwen3-Coder-Next"><b>Qwen/Qwen3-Coder-Next</b></a> β higher-bit GGUFs run on this same fork. We optimize for throughput in ROCmFP4; if you want the last bit of fidelity over speed, that's the one to grab.
|
| 189 |
+
</div>
|
| 190 |
|
| 191 |
**Fast even without speculative decoding.** 3B active params + linear Gated-DeltaNet attention β ~54 t/s short-context decode on a Ryzen AI Max+ 395 (Vulkan0), and cheap long context. No MTP needed.
|
| 192 |
|
|
|
|
| 228 |
# code-weighted imatrix on the Q8 (single pass; ratio = the real lever)
|
| 229 |
llama-imatrix -m Qwen3-Coder-Next-Q8_0.gguf -f code-weighted-calib.txt -o coder-next.imatrix -c 512 -ngl 999
|
| 230 |
|
| 231 |
+
# quant -> ROCmFP4 with the imatrix (Q8 embeddings) + Q6 output head β the β
file (Β§01)
|
| 232 |
+
# fast single-scale body; --output-tensor-type q6_K raises the output head to Q6_K
|
|
|
|
|
|
|
|
|
|
| 233 |
llama-quantize --allow-requantize --token-embedding-type q8_0 --output-tensor-type q6_K --imatrix coder-next.imatrix \
|
| 234 |
+
Qwen3-Coder-Next-Q8_0.gguf Qwen3-Coder-Next-ROCmFP4-STRIX-embQ8-imatrix-headQ6.gguf Q4_0_ROCMFP4_STRIX
|
| 235 |
```
|
| 236 |
|
| 237 |
> Experimental research build for AMD Strix Halo β hardware/driver/prompt-sensitive, may not reproduce elsewhere. Not native FP4 tensor-core execution.
|