add .dockerignore

main.py

@@ -21,11 +21,25 @@ TASK_TYPE = "text-generation"
 
 TASKS = [
     {
-        "label": "OSP-1.4B-100B-Muon-SSNorm-EmbProj",
+        "label": "DRPO-1.5B-Lambda-0.2 @ hygon",
-        "model": "https://www.modelscope.cn/models/dmis-lab/OSP-1.4B-100B-Muon-SSNorm-EmbProj",
+        "model": "https://huggingface.co/mradermacher/DRPO-1.5B-Lambda-0.2-i1-GGUF",
-        "gpu": "Biren_166m",
+        "gpu": "hygon_k100-ai",
-        "framework": "vllm_fix_tokenizer",
+        "framework": "llamacpp",
-        "config": "framework: vllm_fix_tokenizer\nlang: en\nmax_model_len: 2048\nsut_config:\n  gpu_num: 1\n  values:\n    command:\n    - /opt/entrypoint.sh    - /model\n    - --port\n    - '80'\n    - --served-model-name\n    - llm\n    - --max-model-len\n    - '2048'\n    - --enforce-eager\n    - --trust-remote-code\n    - -tp\n    - '1'\nref_config:\n  gpu_num: 1\n  values:\n    command:\n    - /opt/entrypoint.sh    - /model\n    - --port\n    - '80'\n    - --served-model-name\n    - llm\n    - --max-model-len\n    - '2048'\n    - --enforce-eager\n    - --trust-remote-code\n    - -tp\n    - '1'"
+        "config": "framework: llamacpp\napi: completion\nmax_tokens: 1024\ntemperature: 0.7\nrepetition_penalty: 1.2\ntop_p: 0.9\nlang: en\nmax_model_len: 4096\nsut_config:\n  gpu_num: 1\n  values:\n    command:\n    - /app/llama-server\n    - --model\n    - /model/DRPO-1.5B-Lambda-0.2.i1-Q4_K_M.gguf\n    - --alias\n    - llm\n    - --threads\n    - '20'\n    - --n-gpu-layers\n    - '999'\n    - --prio\n    - '3'\n    - --min_p\n    - '0.01'\n    - --ctx-size\n    - '2048'\n    - --host\n    - 0.0.0.0\n    - --port\n    - '8000'\n    - --jinja\n    - --flash-attn\n    - 'off'\nref_config:\n  gpu_num: 1\n  values:\n    command:\n    - /workspace/llama.cpp/build/bin/llama-server\n    - --model\n    - /model/DRPO-1.5B-Lambda-0.2.i1-Q4_K_M.gguf\n    - --alias\n    - llm\n    - --threads\n    - '20'\n    - --n-gpu-layers\n    - '128'\n    - --ctx-size\n    - '4096'\n    - --host\n    - 0.0.0.0\n    - --port\n    - '8000'\n    - --jinja\n"
+    },
+    {
+        "label": "Llama-3.2-1B-abliterated2 @ S4000",
+        "model": "https://huggingface.co/mradermacher/Llama-3.2-1B-Instruct-abliterated2-i1-GGUF",
+        "gpu": "Mthreads_s4000",
+        "framework": "llamacpp",
+        "config": "framework: llamacpp\napi: completion\nmax_tokens: 1024\nmax_model_len: 2048\nsut_config:\n  gpu_num: 1\n  values:\n    command:\n    - /app/llama-server\n    - --model\n    - /model/Llama-3.2-1B-Instruct-abliterated2.i1-Q4_K_M.gguf\n    - --alias\n    - llm\n    - --threads\n    - '20'\n    - --n-gpu-layers\n    - '128'\n    - --ctx-size\n    - '2048'\n    - --host\n    - 0.0.0.0\n    - --port\n    - '8000'\n    - --jinja\n    - --flash-attn\n    - 'off'\nref_config:\n  gpu_num: 1\n  values:\n    command:\n    - /workspace/llama.cpp/build/bin/llama-server\n    - --model\n    - /model/Llama-3.2-1B-Instruct-abliterated2.i1-Q4_K_M.gguf\n    - --alias\n    - llm\n    - --threads\n    - '20'\n    - --n-gpu-layers\n    - '128'\n    - --ctx-size\n    - '2048'\n    - --host\n    - 0.0.0.0\n    - --port\n    - '8000'\n    - --jinja\n    - --flash-attn\n    - 'off'\n"
+    },
+    {
+        "label": "Llama-3.1-SuperNova-Lite @ hygon",
+        "model": "https://huggingface.co/mradermacher/Llama-3.1-SuperNova-Lite-GGUF",
+        "gpu": "hygon_k100-ai",
+        "framework": "llamacpp",
+        "config": "framework: llamacpp\napi: completion\nmax_tokens: 1024\ntemperature: 0.7\nrepetition_penalty: 1.2\ntop_p: 0.9\nlang: en\nmax_model_len: 4096\nsut_config:\n  gpu_num: 1\n  values:\n    command:\n    - /app/llama-server\n    - --model\n    - /model/Llama-3.1-SuperNova-Lite.Q4_K_M.gguf\n    - --alias\n    - llm\n    - --threads\n    - '20'\n    - --n-gpu-layers\n    - '999'\n    - --prio\n    - '3'\n    - --min_p\n    - '0.01'\n    - --ctx-size\n    - '2048'\n    - --host\n    - 0.0.0.0\n    - --port\n    - '8000'\n    - --jinja\n    - --flash-attn\n    - 'off'\nref_config:\n  gpu_num: 1\n  values:\n    command:\n    - /workspace/llama.cpp/build/bin/llama-server\n    - --model\n    - /model/Llama-3.1-SuperNova-Lite.Q4_K_M.gguf\n    - --alias\n    - llm\n    - --threads\n    - '20'\n    - --n-gpu-layers\n    - '128'\n    - --ctx-size\n    - '4096'\n    - --host\n    - 0.0.0.0\n    - --port\n    - '8000'\n    - --jinja\n"
     }
 ]