Lab release history
Last updated Apr 22, 2026
Tencent Hunyuan model releases
Tencent's Hunyuan foundation-model team, releasing large MoE and multimodal models. This page collects the lab's model releases, lifecycle events, source links, and model metadata in one crawlable record.
4
Models
1
Labs
2
Open
4
Recent
4 models
Hunyuan-A13B-Instruct
AvailableTencent HunyuanOpen weights
Tencent Hunyuan open-weight fine-grained MoE model with 80B total parameters and 13B active parameters, optimized for agentic tool use.
Type
MoE
Params
80B
Context
—
Top eval
—
Released
Apr 22, 2026
2mo ago
MoE80B— ctxApr 22, 2026
Hunyuan-Large
AvailableTencent HunyuanOpen weights
Tencent's 389B total / 52B active open-weight Transformer MoE, released with a 256K pretraining context and 128K instruct context.
Type
MoE
Params
389B
Context
128K
Top eval
—
Released
Nov 4, 2024
1.6y ago
MoE389B128K ctxNov 4, 2024
Hunyuan Turbo
AvailableTencent HunyuanProprietary
Tencent's faster, lower-cost Hunyuan update before the open Hunyuan-Large model card.
Type
—
Params
Undisc.
Context
—
Top eval
—
Released
Sep 5, 2024
1.8y ago
—Undisc.— ctxSep 5, 2024
Hunyuan
AvailableTencent HunyuanProprietary
Tencent's first Hunyuan foundation model release, introduced as a general-purpose Chinese enterprise model.
Type
—
Params
Undisc.
Context
—
Top eval
—
Released
Sep 7, 2023
2.8y ago
—Undisc.— ctxSep 7, 2023