Source files / install/strata · raw
flash-next-p100.json
{ "exe": "/opt/tersler/src/p100-strata/strata/build-p100/strata", "args": [ "--pack", "/opt/tersler/strata/packs/flash-next-iq3s", "--native", "/opt/tersler/models/flash-next/Qwen3.8-Flash-Next-GSQ-RCO-IQ3_S-00001-of-00002.gguf", "--ple-gguf", "/opt/tersler/models/flash-next/Qwen3.8-Flash-Next-GSQ-RCO-IQ3_S-00002-of-00002.gguf", "--ple-io", "direct", "--expert-profile", "/opt/tersler/src/p100-strata/strata/data/expert-profile.bin", "--expert-cache", "auto", "--prefill", "auto", "--spec", "4", "--spec-min-p", "0.5", "--mtp", "/opt/tersler/strata/mtp/rt", "--max-context", "262144", "--kv", "int8", "--kv-resident", "32768", "--vram-reserve-later-mib", "1103", "--peer-device", "1", "--peer-reserve-mib", "2000", "--vision" ], "gpu": [ 0, 1 ], "layer_split": "auto", "cwd": "/opt/tersler/src/p100-strata/strata", "tokenizer": "/opt/tersler/strata/packs/flash-next-iq3s/tokenizer", "model_name": "qwen3.8-flash-next-iq3_s-strata", "log": "/opt/tersler/strata/strata.log", "host": "0.0.0.0", "port": 8080, "summary_reasoning_budget_tokens": 1024, "summary_prompt_prefixes": [ "You are a context summarization assistant." ], "summary_min_tokens": 40960, "env": { "STRATA_ROUTE_TAIL_SKIP": "0", "STRATA_Q8_SM60": "1", "STRATA_GU_F16": "2", "STRATA_SM60_D20_SUB": "1", "STRATA_DOWN_F16": "1", "STRATA_B6_F16": "1" }, "vision": { "exe": "/opt/tersler/src/p100-strata/strata/build-vision/bin/strata-vision", "mmproj": "/opt/tersler/models/flash-next/mmproj-Qwen3.8-Flash-Next-BF16.gguf", "model": "/opt/tersler/models/flash-next/Qwen3.8-Flash-Next-GSQ-RCO-IQ3_S-00001-of-00002.gguf", "threads": 12 } }