inference/research
verifiedhigh · 0.90updated 2026-10-04

Qwen/Qwen3-8B config.json

repo

Fields

id
sources/mbl-cfg-qwen3-8b
Publisher
Hugging Face
Kind
repo
Accessed
2026-10-04
notes
num_hidden_layers 36, hidden_size 4096, num_attention_heads 32, num_key_value_heads 8, head_dim 128, rope_theta 1000000, rope_scaling null, max_position_embeddings 40960. Read raw 2026-10-04.

Notes

num_hidden_layers 36, hidden_size 4096, num_attention_heads 32, num_key_value_heads 8, head_dim 128, rope_theta 1000000, rope_scaling null, max_position_embeddings 40960. Read raw 2026-10-04.

Cross-references

This record references no other records.