Qwen3-30B-A3B-4bit

Revision d388dead1515f5e085ef7a0431dd8fadf0886c57

Open on Hugging Face Plan a deployment

30.5BTOTAL PARAMETERS

—CHECKPOINT BYTES

128 / 8EXPERTS / PER TOKEN

7KDOWNLOADS

TENSOR ACCOUNTING

Where the bytes are.

Summed from the safetensors index, one row per dtype. A parameter count alone cannot produce this figure, because a checkpoint may mix widths.

Dtype	Parameters	Bytes each	Bytes	Share
`U32`	30,531,911,680	4	122.1 GB	—
`BF16`	210,944	2	421888 B	—

EVERY FIELD, WITH ITS ORIGIN

Sourced or undetermined. Never assumed.

Each value below names the exact API field it was computed from. Where the Hub does not establish a value, the reason is shown instead of a plausible default.

Architecture: qwen3_moeconfig.model_type
Model classes: Qwen3MoeForCausalLMconfig.architectures
Routed experts: 128config.num_experts
Experts per token: 8config.num_experts_per_tok
Shared experts: UndeterminedThe published config declares no always-on shared experts.
Routing sparsity: 16.0× (1 of every 16.0 experts)config.num_experts / config.num_experts_per_tok
Total parameters: 30,532,122,624safetensors.total
Checkpoint bytes: UndeterminedThis checkpoint packs 4-bit weights into wider integer containers, so element count multiplied by container width is not the checkpoint size. Use repository storage, which is measured on disk.
Ships below 16-bit: Yesconfig.quantization_config.bits
Quantisation method: UndeterminedThe repository declares no quantization method, which normally means unquantized weights.
Trained context: UndeterminedThe config summary omits max_position_embeddings. Trained context is a model-card claim, not a derivable fact.
Declared licence: apache-2.0cardData.license
Base model: Qwen/Qwen3-30B-A3BcardData.base_model
Library: mlxlibrary_name
Files in repository: 14siblings
Last modified: 2025-04-29lastModified

LICENCE POSTURE

Permissive

The repository declares a licence that is generally read as permitting commercial use. Read the licence file in the repository before relying on that.

This is a reading of a metadata field, not legal advice, and it does not account for the licences of upstream models or training data.

PROVENANCE

Derived from https://huggingface.co/api/models/mlx-community/Qwen3-30B-A3B-4bit in the snapshot generated 2026-09-05. Evidence class hub_derived: computed mechanically from publisher metadata, never measured by this project.

Back to catalogue How evidence is classified