This standard quant uses the original SenseTime weights at fa381f7c466739475e172a7d3234e77351e4f613.
It retains a BF16 output head, the original native 524,288-token position encoding, and
144 FP32 recurrent and router tensors. The input embedding, vision linears and native draft
linears are 8-bit. Exact allocations and serialized storage are in PRECISION-ALLOCATION.csv,
TENSOR-INVENTORY.csv and quantization_config.json.
The release was checked with two independent GPU loads, text and vision fixtures, and four target-only streams using a shared 262,144-token Q6 cache. The sampled target-only peaks were 15.31 and 14.59 GiB in the two test shapes. The native speculative weights are included. Exact speculative parity and a physical 16 GiB deployment are separate, unfinished checks. Use target-only mode for the verified setting.
Use the serving source, reference profile, and serving guide. The profile includes the measured residual and router paths. The guide describes the selected settings and throughput for this quant.
model-00001-of-00002.safetensors 4.3 GBmodel-00002-of-00002.safetensors 4.0 GB