{"status": "ok", "model": "Qwen3.8-Flash-Next", "model_ids": ["Qwen3.8-Flash-Next"], "max_batch_size": 5, "warming": false, "memory": {"active": 81522473376, "cache": 0, "peak": 81522473376}, "live": {"connections": 0, "waiting": 0, "decode_tokens_per_second": 0.0, "prefill_tokens_per_second": 0.0}, "version": "1.1.4", "backend": "cuda", "device": "NVIDIA GB10"}