Hippius
v20s550safetensorsbf16Public

mattts/teutonic-q3-8b-5fbywvfq-v20s550

sha256:b5b1611e2ddce49fa53d854470a667ab1fc3f10faa70893e19c357f6a2399bd9·Indexed May 26, 2026

Parameters

8.2B

Total size

15.3 GB

Files

10

Quantization

BF16

No README on this version

Push a README.md with the model files to see it rendered here.

Model architecture

config.json
Architecture
Qwen3ForCausalLM
Model type
qwen3
Hidden size
4,096
Layers
36
Attention heads
32 (8 kv)
FFN size
12,288
Vocab size
151,936
Context window
40K
Dtype
bfloat16

Files

10 items
  • model-00001-of-00005.safetensors

    bfe44a0c4e70

    3.7 GB

    safetensors

  • model-00003-of-00005.safetensors

    1be5f2d89c9d

    3.7 GB

    safetensors

  • model-00004-of-00005.safetensors

    d991116e8412

    3.7 GB

    safetensors

  • model-00002-of-00005.safetensors

    dda008270bb0

    3.6 GB

    safetensors

  • model-00005-of-00005.safetensors

    39198a9aa414

    544.0 MB

    safetensors

  • tokenizer.json

    be75606093db

    10.9 MB

  • model.safetensors.index.json

    61271c4a37be

    32.1 KB

  • config.json

    f7c4eadfbbf5

    728 B

  • tokenizer_config.json

    2e3bd3cb5055

    693 B

  • generation_config.json

    974a56ba6712

    213 B