Image-Text-to-Text
GGUF
German
English
llama.cpp
mtp
speculative-decoding
qwen3
multimodal
conversational
Instructions to use Davidmg0815/Qwen3.8-27B-MTP-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 Davidmg0815/Qwen3.8-27B-MTP-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 Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0 # Run inference directly in the terminal: llama cli -hf Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0 # Run inference directly in the terminal: llama cli -hf Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0 # Run inference directly in the terminal: ./llama-cli -hf Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0 # Run inference directly in the terminal: ./build/bin/llama-cli -hf Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
Use Docker
docker model run hf.co/Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
- LM Studio
- Jan
- vLLM
How to use Davidmg0815/Qwen3.8-27B-MTP-GGUF with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Davidmg0815/Qwen3.8-27B-MTP-GGUF" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Davidmg0815/Qwen3.8-27B-MTP-GGUF", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
- Ollama
How to use Davidmg0815/Qwen3.8-27B-MTP-GGUF with Ollama:
ollama run hf.co/Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
- Unsloth Studio
How to use Davidmg0815/Qwen3.8-27B-MTP-GGUF with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for Davidmg0815/Qwen3.8-27B-MTP-GGUF to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for Davidmg0815/Qwen3.8-27B-MTP-GGUF to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for Davidmg0815/Qwen3.8-27B-MTP-GGUF to start chatting
- Pi
How to use Davidmg0815/Qwen3.8-27B-MTP-GGUF with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
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:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use Davidmg0815/Qwen3.8-27B-MTP-GGUF with Docker Model Runner:
docker model run hf.co/Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
- Lemonade
How to use Davidmg0815/Qwen3.8-27B-MTP-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
Run and chat with the model
lemonade run user.Qwen3.8-27B-MTP-GGUF-Q8_0
List all available models
lemonade list
- Hermes Agent
How to use Davidmg0815/Qwen3.8-27B-MTP-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 Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
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://127.0.0.1:8080/v1 hermes config set model.default Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use Davidmg0815/Qwen3.8-27B-MTP-GGUF with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0
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://127.0.0.1:8080/v1 \ --custom-model-id "Davidmg0815/Qwen3.8-27B-MTP-GGUF:Q8_0" \ --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"
File size: 5,442 Bytes
04b149c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 | #!/usr/bin/env python3
"""MTP-Messreihe gegen einen llama-server. Deutsche Prompts, nach Textsorte
getrennt, weil die Draft-Trefferquote genau daran haengt.
Der Server meldet pro Anfrage draft_n / draft_n_accepted in `timings` — daraus
kommen die Zahlen, nichts wird geschaetzt. Jeder Prompt laeuft mehrfach, ausgewertet
wird der Median, weil Einzelmessungen um +/-6 Prozent schwanken.
./bench_mtp.py --port 8270 --runs 3 --out ergebnis.md
"""
import argparse
import json
import statistics
import sys
import time
import urllib.request
# Textsorten, bei denen MTP erfahrungsgemaess sehr unterschiedlich trifft.
PROMPTS = [
("Aufzaehlung",
"Zaehle von 1 bis 60, nur die Zahlen mit Komma getrennt, sonst nichts."),
("Struktur/JSON",
"Gib ein JSON-Objekt mit 12 deutschen Staedten aus, je mit den Feldern "
"name, bundesland, einwohner. Nur das JSON, kein Text drumherum."),
("Code",
"Schreibe eine Python-Funktion, die eine CSV-Datei einliest, nach der "
"dritten Spalte sortiert und als JSON zurueckgibt. Nur Code."),
("Fliesstext",
"Schreibe drei Absaetze darueber, warum der Rhein fuer die deutsche "
"Industriegeschichte wichtig war."),
("Fachtext",
"Erklaere in etwa 300 Woertern den Unterschied zwischen einem "
"Asynchronmotor und einem Synchronmotor."),
("Kreativ",
"Schreibe eine kurze Geschichte ueber einen Leuchtturmwaerter, der eine "
"Flaschenpost findet."),
("Uebersetzung",
"Uebersetze ins Deutsche: 'The quick assessment of the situation led the "
"engineers to redesign the cooling system, which had failed twice during "
"the summer of the previous year.'"),
("Wiederholung",
"Wiederhole den Satz 'Die Maschine laeuft stabil.' genau 25 Mal, "
"jeweils in einer neuen Zeile."),
]
def ask(port, prompt, max_tokens, timeout):
body = json.dumps({
"messages": [{"role": "user", "content": prompt}],
"max_tokens": max_tokens,
"enable_thinking": False,
"cache_prompt": False,
"temperature": 0.3,
}).encode()
req = urllib.request.Request(
f"http://127.0.0.1:{port}/v1/chat/completions",
data=body, headers={"Content-Type": "application/json"})
t0 = time.time()
with urllib.request.urlopen(req, timeout=timeout) as r:
d = json.load(r)
wall = time.time() - t0
t = d.get("timings", {})
return {
"tps": t.get("predicted_per_second"),
"n": t.get("predicted_n"),
"draft_n": t.get("draft_n") or 0,
"draft_acc": t.get("draft_n_accepted") or 0,
"wall": wall,
}
def main():
ap = argparse.ArgumentParser()
ap.add_argument("--port", type=int, default=8270)
ap.add_argument("--runs", type=int, default=3)
ap.add_argument("--max-tokens", type=int, default=400)
ap.add_argument("--timeout", type=int, default=600)
ap.add_argument("--label", default="")
ap.add_argument("--out", default="")
args = ap.parse_args()
rows = []
for name, prompt in PROMPTS:
tps, accs, drafted, accepted = [], [], 0, 0
for i in range(args.runs):
try:
r = ask(args.port, prompt, args.max_tokens, args.timeout)
except Exception as e:
print(f" {name} Lauf {i+1}: FEHLER {e}", file=sys.stderr)
continue
if r["tps"]:
tps.append(r["tps"])
drafted += r["draft_n"]
accepted += r["draft_acc"]
if r["draft_n"]:
accs.append(r["draft_acc"] / r["draft_n"])
print(f" {name} Lauf {i+1}: {r['tps']:.1f} t/s, "
f"{r['draft_acc']}/{r['draft_n']} Draft-Treffer", file=sys.stderr)
if not tps:
continue
rows.append({
"sorte": name,
"tps": statistics.median(tps),
"tps_min": min(tps),
"tps_max": max(tps),
"quote": (accepted / drafted) if drafted else 0.0,
"drafted": drafted,
"accepted": accepted,
})
hdr = f"### Messreihe {args.label}\n\n" if args.label else "### Messreihe\n\n"
md = hdr
md += "| Textsorte | t/s (Median) | Spanne | Draft-Trefferquote | angenommen / erzeugt |\n"
md += "|---|---|---|---|---|\n"
for r in rows:
md += (f"| {r['sorte']} | **{r['tps']:.1f}** | {r['tps_min']:.1f}–{r['tps_max']:.1f} "
f"| {r['quote']*100:.1f} % | {r['accepted']} / {r['drafted']} |\n")
if rows:
gd = sum(r["drafted"] for r in rows)
ga = sum(r["accepted"] for r in rows)
if gd:
md += (f"\nGesamt: **{ga/gd*100:.1f} %** Trefferquote "
f"({ga} / {gd} Draft-Token), {args.runs} Laeufe je Textsorte, "
f"max_tokens={args.max_tokens}.\n")
else:
# Lauf ohne Draft-Modell: es gibt nichts zu treffen.
tps_all = [r["tps"] for r in rows]
md += (f"\nOhne Spekulation — keine Draft-Token. "
f"Median ueber alle Textsorten: **{statistics.median(tps_all):.1f} t/s** "
f"(Spanne {min(tps_all):.1f}–{max(tps_all):.1f}), "
f"{args.runs} Laeufe je Textsorte, max_tokens={args.max_tokens}.\n")
print(md)
if args.out:
open(args.out, "w").write(md)
print(f"geschrieben: {args.out}", file=sys.stderr)
if __name__ == "__main__":
main()
|