HR Briefing

HR Technology

Tencent Cuts AI Token Prices by Up to 92% with Hy4 Flash Launch

Tencent's Hy4 flash cuts token prices up to 92% versus full Hy4, with a one-million-token context window. Vendor tests show task costs at one-twentieth of Hy4, pending independent verification.

Tencent Launches Hy4 Flash as a Lower-Cost Multimodal Model
Tencent Launches Hy4 Flash as a Lower-Cost Multimodal ModelAkela999 / pixabay

Tom Whitfield

· 4 min read

Tencent has cut input token prices by roughly 92% and output prices by about 89% with the launch of Hy4 flash, a lower-cost multimodal AI model released on November 16, 2026. For HR and business-operations teams weighing generative AI tooling, the pricing move touches a direct decision: high-volume document, slide and spreadsheet work can now run at a small fraction of last month's cost on the same vendor's platform.

According to Tencent Cloud's price list, Hy4 flash costs $0.070 per million input tokens and $0.281 per million output tokens, against $0.834 and $2.501 for full Hy4. Cached input costs $0.003 per million tokens, compared with $0.042 for Hy4. The company positions the new model for everyday productivity, while full Hy4 remains the recommendation for complex, long-horizon software engineering.

What does the cheaper model give up?

Hy4 flash carries 224 billion total parameters, including 64 billion Engram parameters, but activates only 12 billion parameters per token. Tencent says its hybrid-attention, LatentMoE and Engram design shrinks the key-value cache — the stored context computations the model reuses — to one twenty-fourth of Hy3's size and one-sixth of Hy4's. Post-training relies on self-distillation from Hy4, with reinforcement learning cutting average end-to-end output tokens by 50%.

On serving, Tencent reports that megakernels, four-bit quantization and DFly push decoding speed beyond 150 tokens per second. These are vendor-reported figures. No independent benchmark appears in the announcement, and buyers should treat the speed claims accordingly.

How does Tencent measure office-task performance?

Rather than leaning on public benchmark scores, Tencent built WorkIndex around typical tasks in its WorkBuddy product, measuring deliverable quality, single-turn output length and task turnaround time. Eight evaluation sets cover slide creation and editing, human-AI slide co-writing, human-AI Markdown co-writing, document editing, spreadsheet repair, audio and video processing, and visual understanding.

The reported results favor the new model on cost. Hy4 flash performs slightly above Hy4 preview overall on WorkIndex, while average cost per task falls to one-thirteenth of Hy4 preview, one-twentieth of Hy4 and one-eighth of DeepSeek-V4.1-Flash-1017 at Max-tier peak-hour pricing. A caveat applies: all comparisons come from Tencent itself.

A separate blind evaluation involved 85 internal WorkBuddy users across 259 tasks in 27 categories. Tencent reports user satisfaction close to Claude Haiku 5.5, along with the shortest end-to-end completion time and the lowest cost per task among the evaluated models. An 85-person internal sample is modest, and the participants were Tencent staff, not outside customers.

The plain takeaway: on Tencent's own tests, the cheap model now wins on cost and speed while matching or beating its predecessor on office deliverables.

When can teams access it, and at what terms?

Hy4 flash is available through Tencent Cloud TokenHub, WorkBuddy, Yuanbao, WeCom, QQ, ima and Marvis, with OpenRouter availability planned. WorkBuddy access is free through November 30, 2026, and TokenHub API usage carries a 50% discount until 12:00 Beijing time that day. The discount covers only usage generated during the promotional period.

For developers, the API identifier is hy4-flash. Tencent Cloud's documentation lists image understanding, reasoning, structured output, function calling and caching, with maximum input and output limits of 960,000 and 64,000 tokens respectively — below the model's one-million-token context window at the input edge. The API supports OpenAI Chat Completions, OpenAI Responses and Anthropic Messages formats, and customers must activate the model service before making calls.

Tencent recommends Hy4 flash for deep search, document processing, slide creation, simple frontend demonstrations and computer use. The company also plans an open-weight release and frequent updates, though it gave no date for either.

What changes for buyers?

The release completes a two-tier lineup that began with full Hy4 on November 4, 2026, which added visual understanding and computer-use capabilities and promised a cost-efficient Flash version. Employers running high-volume productivity and agent work now have a clear low-cost option; those with difficult software projects still need the pricier tier.

Whether Tencent's internal cost and speed results carry over to independent workloads — and to the planned open-weight release — will determine how far the efficiency gains extend beyond Tencent's own services.

Source: unite.ai

Share this article:

More from Tom Whitfield

Tom Whitfield

Show full bio

Market editor covering money, funding and market shifts around Recruiting and talent acquisition, learning and development at HR Briefing.

279 articles

© 2026 HR Briefing. All rights reserved.