Tencent: Hy3

byTencent
Official docs

Hy3 is a reasoning model from Tencent released on Jul 6, 2026. It reads text and writes text, with a 256K token context window and up to 128K tokens per reply.

Show more

It costs $0.14 per million input tokens and $0.58 per million output tokens. Cached input bills at $0.035, 75% less than fresh input.

It supports tool calling, structured JSON output and prompt caching, and its weights are open, so you can host it yourself.

17 providers serve it, including SiliconFlow, Hugging Face and Novita AI. The lowest input price is $0.066 per million tokens at NanoGPT.

On the Artificial Analysis Intelligence Index it scores 25.3, which places it at number 26 among the 43 models I track with independent scores.

Modalities
In / out price
$0.14 / $0.58 per 1M
Context
256K
Max output
128K
Released
Jul 6, 2026
Knowledge cutoff
n/a
  • Tool calling
  • Structured output
  • Reasoning
  • Prompt caching
  • Open weights
  • Free to use

Hy3 pricing

USD, as listed on Nov 18, 2026. Most bills are input heavy, so the cached input rate matters as much as the headline price.

Input$0.14 / 1M tokens
Output$0.58 / 1M tokens
Cached input (read)$0.035 / 1M tokens

Hy3 providers

17 providers serve it, including SiliconFlow, Hugging Face and Novita AI. The lowest input price is $0.066 per million tokens at NanoGPT. Each host sets its own price and limits, so the cheapest one is not always the one with the longest context.

ProviderInput /MOutput /MCache read /MContextMax output
SiliconFlow$0.132$0.528$0.033262K262K
Hugging Face$0.14$0.58n/a262K128K
Novita AI$0.14$0.58$0.035262K262K
Vercel AI Gateway$0.14$0.58$0.035262K262K
NanoGPT$0.066$0.26$0.029262K128K
DevPass (LLM Gateway)$0.132$0.528$0.033262K128K
Eden AI$0.132$0.528$0.033256K128K
Kilo Gateway$0.132$0.528$0.033262K128K
LLM Gateway$0.132$0.528$0.033262K131K
CommandCode$0.14$0.58$0.035256K128K
Requesty$0.14$0.58$0.035262K262K
Kenari$0.1543$0.6286$0.0154197K128K

Showing 12 of 17 providers: the lab first, then the major clouds, then the cheapest resellers.

What Hy3 costs per month

Pick a workload close to yours or type your own volumes.

Hy3 per month
$11.87
Without caching
$16.59
Per request
$0.0004

Independent benchmarks

Measured by third parties, not reported by Tencent. Each links to its source and shows the date I checked it.

Supported request parameters

What you can set when you call Hy3 through an OpenAI compatible API.

  • Toolstools
  • Structured outputsstructured_outputs
  • Reasoningreasoning
  • Temperaturetemperature

Alternatives to Hy3

Models that take the same inputs and call tools, closest in price, cheaper ones first.

  • Index 39.8

    Reads text, images and video. Tool calling, structured output, reasoning. Served by 31 providers.

    by qwenAug 26, 20261M context$0.15/M input tokens$0.47/M output tokens31 providers

  • Reads text and images. Tool calling, structured output, reasoning. Served by 44 providers.

    by googleApr 2, 2026262K context$0.14/M input tokens$0.40/M output tokens44 providers

  • Reads text, images and files. Tool calling, structured output, reasoning. Served by 45 providers.

    by anthropicOct 7, 20261M context$0.10/M input tokens$0.50/M output tokens45 providers

  • Reads text, images and files. Tool calling, structured output, reasoning. Served by 46 providers.

    by openaiSep 22, 20261.05M context$0.10/M input tokens$0.50/M output tokens46 providers

  • Reads text and images. Tool calling, structured output, reasoning. Served by 32 providers.

    by googleApr 2, 2026$0.13/M input tokens$0.40/M output tokens32 providers

  • Reads text, images, files and video. Tool calling, structured output, reasoning. Served by 120 providers.

    by z-aiAug 26, 20261M context$0.1195/M input tokens$0.4183/M output tokens120 providers

More from Tencent

  • Reads text and images. Tool calling, structured output, reasoning. Served by 7 providers.

    by tencentNov 16, 2026$0.07/M input tokens$0.281/M output tokens7 providers

  • Reads text and images. Tool calling, structured output, reasoning. Served by 12 providers.

    by tencentNov 4, 2026$0.834/M input tokens$2.50/M output tokens12 providers

  • Reads text and images. Tool calling, reasoning.

    by tencentOct 17, 2026$0.15/M input tokens$0.60/M output tokens

  • Reads text and images. Tool calling, structured output, reasoning.

    by tencentSep 18, 2026$0.37/M input tokens$1.25/M output tokens

  • Reads text. Tool calling, structured output, reasoning. Served by 17 providers.

    by tencentAug 28, 2026$0.834/M input tokens$2.50/M output tokens17 providers

  • Reads text. Served by 2 providers.

    by tencentJun 12, 2026$0.044/M input tokens$0.177/M output tokens2 providers

All Tencent models

Hy3 questions

How much does the Hy3 API cost?

$0.14 per million input tokens and $0.58 per million output tokens, with cached input at $0.035. The estimator on this page turns that into a monthly bill for your own workload.

What is the context window of Hy3?

256,000 tokens of input, and it can write up to 128,000 tokens in one reply. As a rough guide, a million tokens is around 750,000 English words.

Which providers serve Hy3?

17 providers serve it, including SiliconFlow, Hugging Face and Novita AI. The lowest input price is $0.066 per million tokens at NanoGPT. Prices differ by host, so the Providers table on this page lists each one with its input and output price.

Does Hy3 support tool calling?

Yes. It accepts tool definitions and can return JSON that matches a schema, which is what an agent needs to call your APIs.

Is Hy3 open source?

Its weights are published, so you can run it on your own hardware or a cloud of your choice. Read the license before commercial use.

When was Hy3 released?

Jul 6, 2026.

How good is Hy3 compared with other models?

It scores 25.3 on the Artificial Analysis Intelligence Index and 1454 on the Arena text leaderboard. Benchmarks are a shortlist, not a verdict: test the top two or three candidates on your own tasks before you commit.

Catalog data from models.dev (MIT licensed), refreshed daily, last on Nov 18, 2026. Independent scores from my hand checked rankings, sources linked above. The summary is written from these facts.