Models·EASYHUB JOURNAL
Hunyuan A13B opens hybrid reasoning with 80B total parameters
What changed
Hunyuan-A13B is a hybrid-reasoning MoE model with 80B total parameters and roughly 13B active per inference. Tencent opens the model and API alongside ArtifactsBench and C3-Bench evaluation datasets. The 13B figure describes active computation, not the full weight size of a 13B dense model.
- Original title
- 业界首个：腾讯混元-A13B 模型发布并开源，极端条件 1 张中低端 GPU 卡即可部署
- Source
- IT之家 · www.ithome.com
- Topic
- Models
- Source month
- 2025-06
This is a concise EasyHub summary of the linked source, not the full report or original reporting. Availability and preview conditions are described in the summary and original.
Summary page published · Editorial information