ModelPrésentéExterne

Bonsai 4b 1bit HNPU

Prebuilt Hexagon NPU artifacts for prism-ml/Bonsai-4B.

Sourcehuggingface.co/runanywhere/bonsai_4b_1bit_HNPU/tree/77f35845509798930c21f23519bab2b627d5afa8

TypeOtherModèle de baseprism-ml/Bonsai-4BVersionRevision 77f3584CClasse

Source du modèle

Type: Other
Taille du fichier: 601–741 MB
Modèle de base: prism-ml/Bonsai-4B
Version: vRevision 77f3584
Mis à jour: 3 oct. 2026
Téléchargements: 0
Tâches: Génération de texte
Source: Hugging Face

Description de la source

Bonsai-4B — Hexagon NPU artifacts

Prebuilt Hexagon NPU artifacts for prism-ml/Bonsai-4B.

See the upstream model card for model behavior, license, intended use, and limitations.

Artifacts are arch-pinned; this repo currently contains a v81/ directory.

Sources

1 source

Hugging FacePrincipalVersion 77f35845509798930c21f23519bab2b627d5afa8

huggingface.co/runanywhere/bonsai_4b_1bit_HNPU/tree/77f35845509798930c21f23519bab2b627d5afa8

Vérifié 3 oct.

Artefacts du modèle

2 artefacts

v81/bonsai4b_embed_f16.bin

bin · 741 MB · SHA-256 046edf4688c9…6395 · Hugging Face

Télécharger

v81/bonsai4b_kernel.bin

bin · 601 MB · SHA-256 0793c03782a0…a4e9 · Hugging Face

Télécharger

Extraits de sources

3 extraits

Readme3 oct.huggingface.co/runanywhere/bonsai_4b_1bit_HNPU

--- license: other tags: - hnpu - hexagon - npu - llm - v81 base_model: prism-ml/Bonsai-4B pipeline_tag: text-generation --- # Bonsai-4B — Hexagon NPU artifacts Prebuilt Hexagon NPU artifacts for [`prism-ml/Bonsai-4B`](https://huggingface.co/prism-ml/Bonsai-4B). See the upstream model card for model behavior, license, intended use, and limitations. Artifacts are arch-pinned; this repo currently contains a `v81/` directory.

Readme17 juil.huggingface.co/runanywhere/bonsai_4b_1bit_HNPU

--- license: other tags: - hnpu - hexagon - npu - v75 - v81 - text-generation - 1-bit base_model: prism-ml/Bonsai-4B --- # Bonsai-4B — 1-bit for the Qualcomm Hexagon NPU (HNPU) A QHexRT-runnable bundle of [prism-ml/Bonsai-4B](https://huggingface.co/prism-ml/Bonsai-4B) at **true 1-bit (Q1_0, ~1.125 bit/weight)** for the Hexagon NPU. Runs the full model on-device via a custom HVX op-package. **Supported arches:** v75 (SM8650), v81 (SM8850). Artifacts are arch-pinned; this repo contains `v75/` and `v81/` directories.

Note17 juil.huggingface.co/runanywhere/bonsai_4b_1bit_HNPU

Source context: 0 downloads · 0 likes · Pipeline text-generation · Repo runanywhere/bonsai_4b_1bit_HNPU