Model chatglm3-6b-32k
| namespace | model name | standby gpu | standby pageable | standby pinned memory | gpu count | vRam (MB) | cpu | memory (MB) | state | revision |
|---|---|---|---|---|---|---|---|---|---|---|
| THUDM | chatglm3-6b-32k | Blob | Blob | Blob | 1 | 13800 | 12.0 | 20000 | Normal | 146 |
Image
Prompt
Sample Rest Call
Pods
| tenant | namespace | pod name | state | require resource | allocated resource |
|---|---|---|---|---|---|
| public | THUDM | public/THUDM/chatglm3-6b-32k/146/933 | Standby | {'CPU': 12000, 'Mem': 20000, 'GPU': {'Type': 'Any', 'Count': 1, 'vRam': 13800}} | {'nodename': 'node3', 'CPU': 12000, 'Mem': 20000, 'GPUType': 'A4000', 'GPUs': {'vRam': 0, 'map': {}, 'slotSize': 0, 'totalSlotCnt': 0}, 'MaxContextPerGPU': 2} |
Func
| { "image": "vllm/vllm-openai:v0.4.2", "commands": [ "--model", "THUDM/chatglm3-6b-32k", "--enforce-eager", "--max-model-len", "1500", "--gpu-memory-utilization", "0.99", "--trust-remote-code" ], "envs": [ [ "LD_LIBRARY_PATH", "/usr/local/lib/python3.12/dist-packages/nvidia/cuda_nvrtc/lib/:$LD_LIBRARY_PATH" ] ], "mounts": [ { "hostpath": "/home/brad/cache", "mountpath": "/root/.cache/huggingface" } ], "endpoint": { "port": 8000, "schema": "Http", "probe": "/health" }, "version": 146, "entrypoint": [], "resources": { "CPU": 12000, "Mem": 20000, "GPU": { "Type": "Any", "Count": 1, "vRam": 13800 } }, "standby": { "gpu": "Blob", "pageable": "Blob", "pinned": "Blob" }, "probe": { "port": 80, "schema": "Http", "probe": "/health" }, "sample_query": { "apiType": "openai", "path": "v1/completions", "prompt": "\u665a\u4e0a\u7761\u4e0d\u7740\u5e94\u8be5\u600e\u4e48\u529e", "body": { "max_tokens": "200", "model": "THUDM/chatglm3-6b-32k", "stream": "true", "temperature": "0" } } } |