# Tencent

Published articles for Tencent.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Tencent Hy4 Preview: 770B Open MoE, Agentic Benchmarks, and Where It Fits in 2026

DevFeed: [Tencent Hy4 Preview: 770B Open MoE, Agentic Benchmarks, and Where It Fits in 2026](<https://devfeed.tech/articles/tencent-hy4-preview-770b-open-moe-agentic-benchmarks-and-where-it-fits-in-2026-56325.md>)

Original publisher: [Read original article](<https://www.developersdigest.tech/blog/tencent-hy4-preview-770b-open-moe-2026>)

Author: Developers Digest

Published: 2026-08-31T00:00:00Z

Content type: article

Language: en

Sources: [Developers Digest](<https://devfeed.tech/sources/developers-digest.md>)

Topics: [Mixture of Experts (MoE)](<https://devfeed.tech/topics/mixture-of-experts-moe.md>), [context](<https://devfeed.tech/topics/context.md>), [deepseek](<https://devfeed.tech/topics/deepseek.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-coding](<https://devfeed.tech/tags/agentic-coding.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [apache](<https://devfeed.tech/tags/apache.md>), [attention](<https://devfeed.tech/tags/attention.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [context](<https://devfeed.tech/tags/context.md>), [deepseek](<https://devfeed.tech/tags/deepseek.md>), [live](<https://devfeed.tech/tags/live.md>), [llm-pricing](<https://devfeed.tech/tags/llm-pricing.md>), [model](<https://devfeed.tech/tags/model.md>), [moe](<https://devfeed.tech/tags/moe.md>), [open](<https://devfeed.tech/tags/open.md>), [open-weights](<https://devfeed.tech/tags/open-weights.md>), [openrouter](<https://devfeed.tech/tags/openrouter.md>), [preview](<https://devfeed.tech/tags/preview.md>), [tencent](<https://devfeed.tech/tags/tencent.md>)

### AI overview

Tencent's Hy4 preview is a 770-billion-parameter open Mixture-of-Experts text model with 49 billion active parameters, Apache 2.0 licensing, a 1-million-token context window, and DeepSeek-style sparse attention. The supplied evidence also reports Terminal-Bench and DeepSWE benchmark scores and OpenRouter pricing.

### Source excerpt

Tencent's Hy4 preview ships 770B total parameters with 49B active under Apache 2.0 - a 1M-context text MoE with DeepSeek-style sparse attention, posted Terminal-Bench 85.4 and DeepSWE 64.3, and an OpenRouter price of $0.834/$2.501. Verified against the model card and the live OpenRouter page on August 31, 2026.

## China's Tencent releases new open-source AI model for coding, research tasks

DevFeed: [China's Tencent releases new open-source AI model for coding, research tasks](<https://devfeed.tech/articles/china-s-tencent-releases-new-open-source-ai-model-for-coding-research-tasks-56439.md>)

Original publisher: [Read original article](<https://indianexpress.com/article/technology/artificial-intelligence/chinas-tencent-releases-new-open-source-ai-model-for-coding-research-tasks-10854166/>)

Author: Reuters

Published: 2026-08-29T03:41:31Z

Content type: news

Language: en

Sources: [Technology | The Indian Express](<https://devfeed.tech/sources/technology-the-indian-express.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Open Source Models & Datasets](<https://devfeed.tech/topics/open-source-models-datasets.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Mixture of Experts (MoE)](<https://devfeed.tech/topics/mixture-of-experts-moe.md>), [Software Engineering](<https://devfeed.tech/topics/software-engineering.md>), [Repository](<https://devfeed.tech/topics/repository.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-model](<https://devfeed.tech/tags/ai-model.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [coding](<https://devfeed.tech/tags/coding.md>), [hugging-face](<https://devfeed.tech/tags/hugging-face.md>), [hy4-preview-model](<https://devfeed.tech/tags/hy4-preview-model.md>), [mixture-of-experts](<https://devfeed.tech/tags/mixture-of-experts.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [open-source-ai](<https://devfeed.tech/tags/open-source-ai.md>), [open-source-ai-models](<https://devfeed.tech/tags/open-source-ai-models.md>), [research](<https://devfeed.tech/tags/research.md>), [technology](<https://devfeed.tech/tags/technology.md>), [technology-artificial-intelligence](<https://devfeed.tech/tags/technology-artificial-intelligence.md>), [tencent](<https://devfeed.tech/tags/tencent.md>), [tencent-codebuddy-workbuddy-ai](<https://devfeed.tech/tags/tencent-codebuddy-workbuddy-ai.md>), [tencent-hunyuan-3-0](<https://devfeed.tech/tags/tencent-hunyuan-3-0.md>), [tencent-hy4-hugging-face](<https://devfeed.tech/tags/tencent-hy4-hugging-face.md>), [tencent-hy4-preview](<https://devfeed.tech/tags/tencent-hy4-preview.md>), [tencent-mixture-of-experts-770b](<https://devfeed.tech/tags/tencent-mixture-of-experts-770b.md>), [tencent-open-source-ai-model](<https://devfeed.tech/tags/tencent-open-source-ai-model.md>), [yao-shunyu-tencent-ai](<https://devfeed.tech/tags/yao-shunyu-tencent-ai.md>)

### AI overview

Tencent released a preview of Hy4, an open-source AI model designed for software engineering, research, and financial analysis. The model uses a mixture-of-experts architecture with 770 billion total parameters, about 49 billion of which are used for each text request.

### Source excerpt

Tencent added that the early-release model can sometimes take longer than necessary to work through complex questions and may over-verify its own answers.

## Tencent Hunyuan Paper Maps Coding-Agent Behavior Across Prompts, Tools, State, Permissions, and Runtime Policy

DevFeed: [Tencent Hunyuan Paper Maps Coding-Agent Behavior Across Prompts, Tools, State, Permissions, and Runtime Policy](<https://devfeed.tech/articles/harness-handbook-shows-the-missing-map-for-coding-agents-56164.md>)

Original publisher: [Read original article](<https://www.developersdigest.tech/blog/harness-handbook-agent-behavior-map>)

Author: Developers Digest

Published: 2026-07-16T00:00:00Z

Content type: article

Language: en

Sources: [Developers Digest](<https://devfeed.tech/sources/developers-digest.md>)

Topics: [code search](<https://devfeed.tech/topics/code-search.md>), [Tool](<https://devfeed.tech/topics/tool.md>), [Authorization](<https://devfeed.tech/topics/authorization.md>), [Prompt Engineering](<https://devfeed.tech/topics/prompt-engineering.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-infrastructure](<https://devfeed.tech/tags/agent-infrastructure.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [code-search](<https://devfeed.tech/tags/code-search.md>), [codex](<https://devfeed.tech/tags/codex.md>), [developer-workflow](<https://devfeed.tech/tags/developer-workflow.md>), [evals](<https://devfeed.tech/tags/evals.md>), [harness](<https://devfeed.tech/tags/harness.md>), [permissions](<https://devfeed.tech/tags/permissions.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [search](<https://devfeed.tech/tags/search.md>), [tencent](<https://devfeed.tech/tags/tencent.md>)

### AI overview

A July 2026 Tencent Hunyuan paper maps coding-agent behavior across prompts, tools, state, permissions, and runtime policy, arguing that code search alone is insufficient.

### Source excerpt

A July 2026 paper from Tencent Hunyuan turns agent harnesses into behavior-level maps. The useful lesson for builders is simple: code search is not enough when one behavior spans prompts, tools, state, permissions, and runtime policy.

## Tencent Hy3: An Open Mixture-of-Experts Model with 295B Parameters

DevFeed: [Tencent Hy3: An Open Mixture-of-Experts Model with 295B Parameters](<https://devfeed.tech/articles/tencent-hy3-a-295b-open-moe-that-punches-above-its-weight-56324.md>)

Original publisher: [Read original article](<https://www.developersdigest.tech/blog/tencent-hy3-open-source-moe-model>)

Author: Developers Digest

Published: 2026-07-10T00:00:00Z

Content type: release

Language: en

Sources: [Developers Digest](<https://devfeed.tech/sources/developers-digest.md>)

Topics: [Mixture of Experts (MoE)](<https://devfeed.tech/topics/mixture-of-experts-moe.md>)

Tags: [ai-models](<https://devfeed.tech/tags/ai-models.md>), [apache](<https://devfeed.tech/tags/apache.md>), [hacker-news](<https://devfeed.tech/tags/hacker-news.md>), [moe](<https://devfeed.tech/tags/moe.md>), [news](<https://devfeed.tech/tags/news.md>), [open](<https://devfeed.tech/tags/open.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [performance](<https://devfeed.tech/tags/performance.md>), [pricing](<https://devfeed.tech/tags/pricing.md>), [tencent](<https://devfeed.tech/tags/tencent.md>), [token](<https://devfeed.tech/tags/token.md>)

### AI overview

Tencent's Hy3 is an Apache 2.0 open Mixture-of-Experts model with 295 billion parameters that activates 21 billion parameters per token. The supplied description says it matches flagship performance at flash-tier pricing.

### Source excerpt

Tencent's Hy3 ships 295B parameters but activates only 21B per token, matching flagship performance at flash-tier pricing under Apache 2.0.