Tencent (Hunyuan)

Tencent Hy4 Preview

Generally availableOpen weights1M contextReasoning

Summary

Tencent Hy4 Preview is an open-weight language model from Tencent (Hunyuan), released on 28 Aug 2026. It accepts text and generates text, with a 1M-token context window and up to 64K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.83 per million input tokens and $2.50 per million output tokens (DeepInfra), across 5 providers we track.

A next-generation productivity model with significantly enhanced Agent and complex task execution capabilities.

Last verified 18 Nov 2026. Every figure on this page links to its source.

Specifications

DeveloperTencent (Hunyuan)
Release date28 Aug 2026 [models.dev]
StatusGenerally available [models.dev]
WeightsOpen weights [Hugging Face]
Licenseapache-2.0 [Hugging Face]
Parameters780B [Hugging Face]
Context window1M (1,024,000 tokens) [models.dev]
Max output64K tokens [models.dev]
InputText [models.dev]
OutputText [models.dev]
Reasoning modeYes [models.dev]
Tool callingYes [models.dev]
Structured outputYes [models.dev]
Hugging Facetencent/Hy4-preview [Hugging Face]

Tencent Hy4 Preview API pricing by provider

USD per million tokens. Sorted by input price.

ProviderInputOutputCached inputContextSource
DeepInfra
tencent/Hy4-preview
$0.83$2.50$0.041Mmodels.dev, 18 Nov 2026
Hugging Face Inference Providers
tencent/Hy4-preview
$0.83$2.50–1Mmodels.dev, 18 Nov 2026
Novita AI
tencent/hy4-preview
$0.83$2.50$0.041Mmodels.dev, 18 Nov 2026
OpenRouter
tencent/hy4-preview
$0.83$2.50$0.041MOpenRouter, 18 Nov 2026
Vercel AI Gateway
tencent/hy4-preview
$0.83$2.50$0.041Mmodels.dev, 18 Nov 2026

Running Tencent Hy4 Preview locally

Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.

QuantizationWeightsTotal, 8K contextTotal, 32K context
Q4_K_M (4-bit)473 GB474 GB477 GB
Q8_0 (8-bit)829 GB830 GB832 GB
FP16 / BF16 (16-bit)1560 GB1561 GB1564 GB

At 4-bit with an 8K conversation, Tencent Hy4 Preview needs about 474 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.

Frequently asked questions

What is the context window of Tencent Hy4 Preview?

Tencent Hy4 Preview has a context window of 1,024,000 tokens (1M) and can generate up to 64,000 tokens in a single response, according to models.dev.

How much does the Tencent Hy4 Preview API cost?

Tencent Hy4 Preview is listed from $0.83 per million input tokens and $2.50 per million output tokens (DeepInfra). Prices last verified 18 Nov 2026.

Is Tencent Hy4 Preview open source?

Tencent Hy4 Preview is an open-weight model: its weights are published on Hugging Face as tencent/Hy4-preview under the apache-2.0 license. Check the license terms before commercial use.

When was Tencent Hy4 Preview released?

Tencent (Hunyuan) released Tencent Hy4 Preview on 28 Aug 2026, according to models.dev.

Which providers offer Tencent Hy4 Preview?

We track Tencent Hy4 Preview on 5 providers: DeepInfra, Hugging Face Inference Providers, Novita AI, OpenRouter, Vercel AI Gateway.

How much VRAM does Tencent Hy4 Preview need?

About 474 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 830 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count and architecture.

Similar models from other labs

Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.

ModelLabReleasedContextInput $/1MOutput $/1MWeights
Step 5 PreviewStepFun16 Sept 20261M$1$2.70Open
Macaron V1 VentiMindlab21 Jul 20261M$0.82$2.48Open
Pareto 26.10 PreviewUnbiased1 Oct 20261M$0.80$3.20–
Nemotron 3 UltraNVIDIA4 Jun 20261M$0.50$2.20Open
Muse Spark OpenMeta15 Oct 20261M$1.25$4.25Open
GLM-5.2Zhipu AI (Z.ai)13 Jun 20261M$0.007$2.40Open

Change history

  • Context window changed from 1M to 1M
  • Name changed from Hy4 preview to Tencent Hy4 Preview
  • OpenRouter price changed: input $0.75 → $0.83, output $2.25 → $2.50 per 1M tokens
  • OpenRouter price changed: input $0.83 → $0.75, output $2.50 → $2.25 per 1M tokens
  • Name changed from Tencent Hy4 Preview to Hy4 preview
  • Context window changed from 1M to 1M
  • Name changed from Hy4 preview to Tencent Hy4 Preview
  • Context window changed from 1M to 1M
  • Name changed from Tencent Hy4 Preview to Hy4 preview
  • Context window changed from 1M to 1M
  • OpenRouter price changed: input $0.75 → $0.83, output $2.25 → $2.50 per 1M tokens
  • Context window changed from 1M to 1M
  • Name changed from Hy4 preview to Tencent Hy4 Preview
  • OpenRouter price changed: input $0.83 → $0.75, output $2.50 → $2.25 per 1M tokens
  • Context window changed from 1M to 1M
  • Name changed from Tencent Hy4 Preview to Hy4 preview
  • Context window changed from 1M to 1M
  • Context window changed from 1M to 1M
  • Name changed from Hy4 preview to Tencent Hy4 Preview
  • Context window changed from 1M to 1M
  • Context window changed from 1M to 1M
  • Context window changed from 1M to 1M
  • Context window changed from 1M to 1M
  • Context window changed from 1M to 1M
  • Context window changed from 1M to 1M

Sources