{ "exe": "strata", "args": [ "--pack", ".", "--native", "./model-skeleton.gguf", "--native-dense-gguf", "./model-skeleton.gguf", "--ple-gguf", "./ple-fp8.gguf", "--ple-io", "direct", "--expert-profile", "./profile.bin", "--expert-cache", "auto", "--prefill", "auto", "--pcie-frac", "0.1", "--spec", "4", "--spec-min-p", "0.5", "--mtp", "./mtp", "--max-context", "131072", "--kv", "int8", "--kv-resident", "0", "--vram-reserve-mib", "700", "--pool-workers", "24", "--pcie-mode", "dma", "--resident-experts", "--no-prefill-borrow", "--conversation-cache-mib", "0" ], "cwd": ".", "tokenizer": "./tokenizer", "model_name": "bench-145b-q3k-skeleton", "log": "./serve-bench.log", "port": 8123, "gpu": 0, "gpus_asked": true, "env": { "STRATA_LATENT_DENSE": "./latent_dense_q8.bin", "STRATA_LATENT_A": "./experts.bin", "STRATA_LATENT_LABELS": "./labels.bin", "STRATA_EMBED_VRAM": "1", "STRATA_LATENT_FUSED": "1", "STRATA_LATENT_FUSED_DU": "1", "STRATA_LATENT_U_GPU": "1", "STRATA_LATENT_DENSE_FMT": "8", "STRATA_LAT_FUSED_Q3K": "1" } }