Platform status
Sampled every 5 minutes; snapshot refreshes every minute.
API gateway
Model scheduler
Updated 20:52 IST
ready
Llama 3.3 70B
llama-3.3-70b
100% · 7 days
ready
Qwen3 Coder 30B
qwen3-coder-30b
100% · 7 days
ready
Qwen2.5-VL 7B Instruct
qwen2.5-vl-7b-instruct
100% · 7 days
ready
Llama 3.1 8B
llama-3.1-8b
100% · 7 days
ready
Qwen3 8B
qwen3-8b
100% · 7 days
ready
BGE-M3
bge-m3
100% · 7 days
ready
MiniLM L6 v2
all-minilm-l6-v2
100% · 7 days
standby means the model is healthy but idle-parked to save GPU memory — the first request wakes it in a few seconds. Only down indicates an actual fault.