Hy3 preview

TencentMCP Tool CallReasoning
tencent/hy3-previewOpen Weights

Tencent Hy reasoning model for coding, instruction following, and agent tasks

Benchmark Pricing$0.17 / 1M in$0.57 / 1M out$0.05 cache read
Context: 256,000 tokens•Max Output: 64,000 tokens•Released: 2026-04-20

TECHNICAL SPECIFICATIONS

Context Window256,000256K tokens
Max Output64,00064K tokens
LicenseOpen SourceSelf-hostable
Input Modalities
Text
Function / Tool CallingSupported (MCP Ready)
Structured OutputsGuaranteed Schema

MODEL CONTEXT PROTOCOL (MCP) READINESS

✓ Hy3 preview natively supports OpenAI / Anthropic format tool calling and is fully compatible with MCP servers and agent loops.

MCP Context Window Recommendations

With a context capacity of 256,000 tokens, this model can handle moderately large MCP tool definitions and multi-turn conversation logs comfortably.

STANDARDIZED BENCHMARKS & EVALS

SWE-Bench Verifiedresolved
74.4

API CALL INVOCATION EXAMPLE

// Example: Calling Hy3 preview via OpenAI SDK
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.TENCENT_API_KEY,
});

const response = await client.chat.completions.create({
  model: "tencent/hy3-preview",
  messages: [{ role: "user", content: "Hello! Can you inspect MCP tools?" }],
});

console.log(response.choices[0].message.content);

PRICING BY PROVIDER

6 providers
Tencent TokenHubhy3-preview
$0.00 / $0.00$0.00 cache
AIHubMixhy3-preview
$0.17 / $0.57$0.05 cache
OpenCodehy3-preview-free
$0.00 / $0.00$0.00 cache
Kilo Gatewaytencent/hy3-preview
$0.18 / $0.60$0.06 cache
ZenMuxtencent/hy3-preview
$0.17 / $0.57$0.06 cache
OpenRoutertencent/hy3-preview
$0.18 / $0.60$0.06 cache

Rates in USD per 1M tokens. Rates synchronized from models.dev public catalog.

Prompt Caching Active

This model supports prompt caching at $0.05 / 1M tokens, saving up to 70% on repetitive MCP tool definitions and system instructions.