hardhant
/

Qwen3.8-Flash-Next-FP4-Local-Strata

Mixture of Experts

Model card Files Files and versions

Qwen3.8-Flash-Next-FP4-Local-Strata

249 GB

Ctrl+K

Ctrl+K

1 contributor

History: 15 commits

hardhant's picture

sizes for all files

50cc704 verified 4 days ago

mtp
MTP draft head (rt) 4 days ago
pack
runtime pack: experts.bin, dense.bin, index, tokenizer 4 days ago
.gitattributes

1.78 kB
vision projector (BF16, unchanged from the base model) 4 days ago
README.md

7.34 kB
README: self-contained release (pack, mtp, mmproj included) 4 days ago
SHA256SUMS

2 kB
checksums for all files 4 days ago
SIZES.txt

871 Bytes
sizes for all files 4 days ago
mmproj-Qwen3.8-Flash-Next-BF16.gguf

908 MB
xet

vision projector (BF16, unchanged from the base model) 4 days ago
ple-fp4local-bf16.gguf

102 GB
xet

PLE n-gram table decoded to BF16 4 days ago
ple-fp4local-bf16.gguf.manifest.json

553 Bytes
add ple-fp4local-bf16.gguf.manifest.json 4 days ago
qwen-fp4local-csf-dense.gguf

73.9 GB
xet

main text GGUF: NVFP4 experts (lossless csf scales) + Q8_0 dense 4 days ago
qwen-fp4local-csf-dense.gguf.manifest.json

553 Bytes
add qwen-fp4local-csf-dense.gguf.manifest.json 4 days ago
token-embd-fp4local-bf16.gguf

1.27 GB
xet

token embedding, BF16 from the checkpoint 4 days ago