Hippius
mainsafetensorsbf16Public

allforone111/albedo-qwen3-4b-refined-v7

sha256:023f0d1630fd19d8c0acba21bc2edb345d2164701150a2a705b0687df69e07c4·Indexed Jun 8, 2026

Parameters

2.0B

Total size

3.8 GB

Files

13

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
2,048
Layers
28
Attention heads
16 (8 kv)
FFN size
6,144
Vocab size
151,936
Context window
40K

Files

13 items
  • model-00001-of-00004.safetensors

    1b7c07c881ca

    1.8 GB

    safetensors

  • model-00002-of-00004.safetensors

    2b69f74b5f89

    676.1 MB

    safetensors

  • model-00003-of-00004.safetensors

    d685637dd41e

    672.1 MB

    safetensors

  • model-00004-of-00004.safetensors

    f34b0cdb7290

    672.1 MB

    safetensors

  • tokenizer.json

    be75606093db

    10.9 MB

  • vocab.json

    ca10d7e9fb3e

    2.6 MB

  • merges.txt

    8831e4f1a044

    1.6 MB

  • model.safetensors.index.json

    2d8c8eb33e10

    25.0 KB

  • tokenizer_config.json

    2a7a9be36543

    9.4 KB

  • chat_template.jinja

    87a2728cb8dc

    4.0 KB

  • config.json

    b851fd5fca71

    1.4 KB

  • special_tokens_map.json

    76862e765266

    613 B

  • generation_config.json

    bcce630cb886

    187 B

allforone111/albedo-qwen3-4b-refined-v7 · main · Hippius Hub