Provider

Tencent

Every Tencent model UltraGPT can reach, kept in sync with the live catalog — pricing, context window and capabilities for each one.

9 models from Tencent Up to 1.05M tokens of context
Open the appBrowse every provider

9

Models tracked

1

Providers

6

Reasoning models

2

See images (vision)

1,049K

Largest context window

0

Free to run

Directory

All Tencent models.

Search and filter this provider's lineup — pricing per million tokens, context window, modalities, reasoning effort and independent benchmark scores where available.

9 of 9 models

Hunyuan-A13B is a 13B active parameter Mixture-of-Experts (MoE) language model developed by Tencent, with a total parameter count of 80B and support for reasoning via Chain-of-Thought. It offers competitive benchmark...

ReasoningStreamingJSON mode
Context window131.1K tok
Max output118.0K tok
ReleasedJul 8, 2025
Knowledge cutoffMar 31, 2025
TextText
Input$0.14 / 1M tok
Output$0.57 / 1M tok

Hy-MT2-1.8B is a compact 1.8B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided...

Streaming
Context window8.2K tok
Max output4.1K tok
ReleasedAug 20, 2026
Knowledge cutoff—
TextText
Input$0.044 / 1M tok
Output$0.177 / 1M tok

Hy-MT2-30B-A3B is Tencent's flagship translation model in the Hy-MT2 family. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and...

StreamingJSON mode
Context window8.2K tok
Max output4.1K tok
ReleasedAug 20, 2026
Knowledge cutoff—
TextText
Input$0.074 / 1M tok
Output$0.295 / 1M tok

Hy-MT2-7B is a 7B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided translation.

StreamingJSON mode
Context window8.2K tok
Max output4.1K tok
ReleasedAug 19, 2026
Knowledge cutoff—
TextText
Input$0.074 / 1M tok
Output$0.295 / 1M tok

Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort:...

ReasoningTool callingStreamingJSON mode
Context window262.1K tok
Max output128K tok
ReleasedJul 6, 2026
Knowledge cutoff—
TextText
Input$0.132 / 1M tok
Output$0.528 / 1M tok
Cache read$0.033 / 1M tok
Effort: high, low, none

Design Arena — ranked #67 across 6 categories

Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use. It supports configurable reasoning levels across disabled, low, and high modes, allowing it to...

ReasoningTool callingStreaming
Context window262.1K tok
Max output235.9K tok
ReleasedApr 22, 2026
Knowledge cutoff—
TextText
Input$0.18 / 1M tok
Output$0.6 / 1M tok
Cache read$0.06 / 1M tok
Effort: high, low, none
Coding58.8
Agentic24.1

The official release of Hy4 delivers comprehensive and substantial improvements over Hy4 Preview, with particularly strong gains in agentic capabilities and performance across a wide range of long-horizon, real-world productivity...

ReasoningTool callingVisionStreamingJSON mode
Context window1.05M tok
Max output65.5K tok
ReleasedNov 4, 2026
Knowledge cutoff—
Text + ImageText
Input$0.417 / 1M tok
Output$1.25 / 1M tok
Cache read$0.021 / 1M tok
Effort: high, low, none
Intelligence34.6

Design Arena — ranked #25 across 5 categories

Hy4 flash is a cost-efficient variant of Hy4 from Tencent, served with roughly a quarter of the flagship's parameter count for lower latency and cost. It targets code assistance, agent...

ReasoningTool callingVisionStreamingJSON mode
Context window1.05M tok
Max output65.5K tok
ReleasedNov 16, 2026
Knowledge cutoff—
Text + ImageText
Input$0.035 / 1M tok
Output$0.14 / 1M tok
Cache read$0.0015 / 1M tok

Design Arena — ranked #46 across 1 categories

Tencent: Hy4 preview is a mixture-of-experts model from Tencent, with 49B active parameters out of 770B total. It is designed for coding agents, complex tool-use workflows, and productivity tasks that...

ReasoningTool callingStreamingJSON mode
Context window1.05M tok
Max output64K tok
ReleasedAug 28, 2026
Knowledge cutoff—
TextText
Input$0.834 / 1M tok
Output$2.5 / 1M tok
Cache read$0.042 / 1M tok
Effort: high, low, none

Every Tencent model above, one subscription away.

UltraGPT puts this whole directory behind a single login and one flat monthly price — no separate bill per provider.