Tencent: Hy3 preview

tencent · Released Apr 22, 2026
DFO score #121 / 744
21.3 leaderboard score
Speed #100 / 353
137.1 tok/s
Input Price #347 / 744
$0.180 per 1M tokens
Output Price #348 / 744
$0.600 per 1M tokens
Context #203 / 744
262,144 tokens

Analysis Summary

Tencent’s Hy3 preview is a text model with a 262K-token context, tool use, and function calling. Its agentic benchmark performance is a notable strength, while the supplied reasoning results are more limited than those of leading general-purpose models. Low listed input and output pricing makes it appealing for experiments and cost-sensitive tool-driven workflows.

Businesses can evaluate Hy3 for multi-step task execution, internal agents, and automation where the model must invoke functions rather than only draft text. Its benchmark profile supports a specialist agent role, but does not establish top-tier performance in general reasoning or client-facing writing. Teams should test tool selection, completion accuracy, and recovery behavior before giving it access to consequential actions.

Consider Hy3 preview for bounded agent pilots where low cost and tool use are priorities. Retain human review and explicit safeguards until the workflow has demonstrated reliable execution.

Assessed September 23, 2026

Editorial notes

Hy3 preview combines strong agentic results and tool support with a 262K context and low listed pricing; its overall reasoning results are more limited than leading general-purpose models.

The DFO score ranks models on the Artificial Analysis Intelligence Index and is refreshed as new models launch. The write-up and verdict are our own. Feedback?

DFO Verdict

Hy3 preview combines strong agentic results and tool support with a 262K context and low listed pricing; its overall reasoning results are more limited than leading general-purpose models.

#121 of 744 overall Down 4 this week

Benchmarks

Intelligence Index 14.5 #120 of 452 · Artificial Analysis
Coding Index Not published yet Artificial Analysis
Agentic Index Not published yet Artificial Analysis
Creative Writing Not rated Arena rating
Instruction Following Not rated Arena rating
GPQA Diamond 86.7%
HLE 25.5%
SciCode 41.2%
TerminalBench Hard 34.1%
τ²-Bench 92.7%
IFBench 63.1%
LCR 54.7%

Artificial Analysis data refreshed Nov 12, 2026. Arena ratings from the Nov 9, 2026 text leaderboard, with style control. Bars: magenta = reasoning, ink = coding and agents, cyan = instructions and long context.

Artificial Analysis has not published coding or agentic results for Tencent: Hy3 preview yet. Until it does, the model sits below tested models on the Coding and AI Agents lists.

Artificial Analysis no longer lists this model, so its indexes are converted from their older scale to stay comparable.

How Tencent: Hy3 preview compares

Tencent: Hy3 preview ranks #120 of 452 AI models we track for overall intelligence. Its 262K-token context window is larger than 73% of the models we list. At $0.18 per million input tokens it is cheaper than 53% of comparable models.

Position in the field
DFO score: ahead of 84% of models #120
Speed: faster than 72% of models #100
Price: cheaper than 53% of models #347
Context: larger than 73% of models #203
worst in fieldmedianbest in field
Price vs frontier peers · $ per 1M tokens
Tencent: Hy3 preview $0.18 in $0.60 out
Anthropic: Claude Fable 5.2 $8.00 in $40.00 out
Google: Gemini 4 Carbon $2.00 in $10.00 out
Google: Gemini 4 Argon $2.00 in $10.00 out

Dark bar = input · light bar = output, scaled to the priciest peer.

Context window vs peers · tokens

1M tokens ≈ 8 full-length novels or ~2,500 pages of business documents in a single request.

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.18 $0.000180
Output $0.60 $0.000600

What would Tencent: Hy3 preview cost your business?

Pick the job that looks most like yours, then fine-tune with the sliders. Estimates update live.

A website chatbot handling around 100 customer conversations a day, a few short messages each.

3,000
One request is one message, email, draft or automation call.
1,200 tokens

$1.10/mo Tencent: Hy3 preview $0.0004 per request
$63.36/mo Anthropic: Claude Fable 5.2 $0.02 per request
$15.84/mo Google: Gemini 4 Carbon · best value $0.0053 per request

Full calculator with 744 models → Price Calculator

DFO AI AUTOMATION

These numbers get smaller with the right architecture.

We route routine calls to cheap models and save Tencent: Hy3 preview for the hard ones. Most clients cut their estimate by 60-80%.

Talk to our team

About Tencent: Hy3 preview

Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use. It supports configurable reasoning levels across disabled, low, and high modes, allowing it to..

Frequently asked questions about Tencent: Hy3 preview

How much does Tencent: Hy3 preview cost?

Tencent: Hy3 preview costs $0.18 per million input tokens and $0.60 per million output tokens.

What is the context window of Tencent: Hy3 preview?

Tencent: Hy3 preview has a context window of 262,144 tokens (262K).

What can Tencent: Hy3 preview do?

Tencent: Hy3 preview supports tool use and function calling.

Who created Tencent: Hy3 preview?

Tencent: Hy3 preview is developed by Tencent and was released on April 22, 2026.

© 2026 Design for Online Ltd. Registered in England and Wales No. 10328553. VAT Registered. Design for Online® and Forerunner® are registered trademarks.