File size: 523 Bytes
1f4c7f2
c776287
1f4c7f2
c776287
 
1f4c7f2
 
 
c776287
1f4c7f2
c776287
 
1f4c7f2
c776287
d28b9f3
1f4c7f2
e2ad9b1
1f4c7f2
d28b9f3
 
1f4c7f2
 
d28b9f3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
#!/bin/bash
set -e

echo "=== Downloading model ==="
MODEL_PATH=$(python3 -c "
from huggingface_hub import hf_hub_download
path = hf_hub_download(
    repo_id='Jackrong/Qwen3.5-9B-Claude-4.6-Opus-Reasoning-Distilled-v2-GGUF',
    filename='Qwen3.5-9B.Q4_K_M.gguf',
)
print(path)
")

echo "Model path: $MODEL_PATH"
echo "=== Starting llama-server on port 7860 ==="

exec /app/llama/llama-server \
    --model "$MODEL_PATH" \
    --host 0.0.0.0 \
    --port 7860 \
    --ctx-size 2048 \
    --threads 2 \
    --batch-size 256