LOADING DATASHEET
GROUNDED IN REAL HUMAN CONVERSATIONSYOUR BENCH
Tencent Hy4
★ NEW
by Tencent
CLOSEDLOCALAPIREASONING
RANKED #177 OF 228 ASSISTANTS · OVERALL #276 OF 6,618 · VIBE SCORE 5.5 · 134 VOICES
A next-generation productivity model with significantly enhanced Agent and complex task execution capabilities.
HAPPY7.6
8 mentions
SAD7.4
8 mentions
VIBE TRAJECTORY
26 weeks · 156 voices
TOO QUIET
Not enough weekly voices to call a trend yet.
PREFER OPEN SOURCE? THE CROWD HAS PICKS
Tencent Hy4 is closed, so you rent it. These open-weight models can genuinely stand in for it. Download once, own forever.
- OPEN WEIGHTS7.5
GLM 5.3ZhipuDirect competitor matching Hy4 in high end coding, reasoning, and productivity.🏆 NEAR-PEER QUALITY💻 STRONG AT CODE❤️ CROWD FAVORITE6.1k VOICESMATCH★★★ - OPEN WEIGHTS7.4
DeepSeek V4 FlashDeepSeekUltra fast and cost efficient MoE model that handles everyday tasks with ease.💸 COST EFFICIENT⚡ FAST🏆 NEAR-PEER QUALITY5.0k VOICESMATCH★★★ - OPEN WEIGHTS6.7DeepSeek R1DeepSeekExceptional deep reasoning and coding power for complex problem solving.💻 STRONG AT CODE🏆 NEAR-PEER QUALITY❤️ CROWD FAVORITE1.9k VOICESMATCH★★★
WHAT THE INTERNET SAYS
People are mostly happy with Tencent Hy4. Biggest praise: help with code.
134 POSTS · 0 COMMENTS · HACKER NEWS · LEMMY · BLUESKY · GITHUB · REDDIT · DEV.TO · BLOG
PEOPLE LOVE
- Popular for coding help (3 mentions)
PEOPLE COMPLAIN ABOUT
- Coding help disappoints some people (1 mentions)
TOPIC BY TOPIC
- Help with codePEOPLE LOVE THIS
3 thumbs up · 1 thumbs down
LOOKING FOR AN ALTERNATIVE?
Other models the crowd has fully reviewed, starting with text models like this one.
> SOURCES (12)
- HACKER NEWS · 2026-08-29 · 70 UPVOTESTencent Releases and Open-Sources Tencent Hy4 Preview
Tencent Releases and Open-Sources Tencent Hy4 Preview
- BLOG · 2026-08-29 · 32 UPVOTESIntroducing Hy4 Preview
Introducing Hy4 Preview New open weight text input (no vision) LLM from Chinese company Tencent today: 770B total parameters, 49B active parameters, 1M token context window, 1.56TB on Hugging Face . This is a big size increase from their previous Hy3 in July,…
- BLOG · 2026-08-29 · 32 UPVOTES[AINews] OpenAI shuts off Cursor
A late entrant in the news cycle of an eventful week: Following the closing of Cursor’s acquisition by SpaceX last week , it was time for OpenAI to do what Anthropic did to Windsurf when it was being considered for acquisition by OpenAI: There are many angles…
- LEMMY · 2026-08-28 · 13 UPVOTESHy4-Preview 770B-A49B preview frontier model from Tencent
Hy4-Preview 770B-A49B preview frontier model from Tencent
- HACKER NEWS · 2026-08-28 · 5 UPVOTESHy4 Preview
Hy4 Preview
- HACKER NEWS · 2026-08-28 · 4 UPVOTESTencent Releases and Open-Sources Tencent Hy4 Preview
Tencent Releases and Open-Sources Tencent Hy4 Preview
- LEMMY · 2026-08-29 · 4 UPVOTESTencent releases Hy4 preview, open-weights AI model. 770B parameters, 49B active, 1M tokens context window.
Tencent releases Hy4 preview, open-weights AI model. 770B parameters, 49B active, 1M tokens context window.
- HACKER NEWS · 2026-08-28 · 3 UPVOTESHy4 Preview
Hy4 Preview
- HACKER NEWS · 2026-08-30 · 3 UPVOTESHy4 Preview
Hy4 Preview
- LEMMY · 2026-08-28 · 2 UPVOTESOpenCode Updated usage/models (28/8/2026)
OpenCode Updated usage/models (28/8/2026)
- REDDIT · 2026-08-29 · 1 UPVOTESTencent compressed Hy4-preview from 1.5TB to about 200GB GGUF and kept about 98% performance.
Tencent compressed Hy4-preview from 1.5TB to about 200GB GGUF and kept about 98% performance.
- GITHUB · 2026-08-28 · 0 UPVOTESfeat(provider-registry): support Hy4, Ling 3.0, and Qwen 3.8 Flash
feat(provider-registry): support Hy4, Ling 3.0, and Qwen 3.8 Flash