Currently listed through these providers:
Model details
Hunyuan A13B Instruct
Hunyuan A13B Instruct is positioned as a general-purpose text-to-text instruction model within Tencent's Hunyuan family, and its emergence was noted by independent AI press in early July 2025. Coverage from a weekly roundup newsletter placed the model in the same release window as Baidu ERNIE 4.5, Pangu Pro MoE, and Agentica DeepSWE, signaling that it was treated by observers as part of a broader wave of Chinese-developed foundation models arriving that week. That independent mention, while not detailed, gives the model a clear place in the mid-2025 model release landscape rather than presenting it as an outlier or experimental artifact.
In practical terms, Hunyuan A13B Instruct is best understood as a flagship-style instruction model from Tencent designed to handle standard chat, reasoning, and structured output tasks through a text-in, text-out interface. The supplied evidence does not corroborate specific architecture choices, parameter counts, training pipelines, or benchmark scores, so any deeper technical characterization would be speculative; instead, the model is best framed as a recent addition to the Hunyuan lineup aimed at developers who want a modern Tencent instruction model accessible through aggregator routing. Readers evaluating it for production use should consult the official Tencent and OpenRouter model documentation for confirmed capability, context, and pricing details.
Quick Info
Powered by- Provider
- OpenRouter
- Model key
- tencent/hunyuan-a13b-instruct
- Release date
- Jul 8, 2025
- Last updated
- Jul 8, 2025
- Knowledge cutoff
- 2025-03-31
- Input modalities
- Output modalities
- Capabilities
Cost
- Input token cost
- $0.14
- Output token cost
- $0.57
Limits
- Output tokens
- 117,964 tokens
- Context window
- 131,072 tokens