How much does the Hy3 API cost?

On FastMetal, Hy3 is billed per token: $0.1477 per 1M input tokens and $0.6119 per 1M output tokens on a US-dollar account, before tax. Usage is drawn from a prepaid balance; there is no subscription or monthly fee.

Can I call Hy3 with the OpenAI SDK?

Yes. Point base_url at https://api.fastmetal.ai/v1 and pass "hy3" as the model; existing OpenAI-style code works unchanged, including streaming, tool calls and structured output.

What is the context window of Hy3?

262,144 tokens, shared between the prompt and the response.

#79 on the public arena's Japanese board (ELO 1,413). Ranks move as the leaderboard is updated.

What do I need to try Hy3?

Create an account and add credit; the model is then available both in the browser chat and over the API. There is no contract or minimum spend.

Back to Models

Tencent

hy3

Hy3

Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort:...

7/6/2026

262,144 tokens

#31 Text (spanish)

Input: $0.1477/M

Output: $0.6119/M

Try in Chat Get API Key

Specifications

Modalities

Input

text

Output

text

Supported Parameters

frequency_penalty

include_reasoning

logit_bias

max_completion_tokens

max_tokens

min_p

presence_penalty

reasoning

reasoning_effort

repetition_penalty

response_format

seed

stop

structured_outputs

temperature

tool_choice

tools

top_k

top_p

Max Output Tokens

262,144

Reasoning Configuration

Default: Thinking on (effort: high)
Selectable effort levels: high
low
none
Turning thinking off: Possible (reasoning.enabled: false)

How to control thinking, and what it costs

Data policy

Prompt retention: Unknown — we could not confirm
Training: Unknown — not declared

"Unknown" does not mean "safe". It means we could not confirm it.

Whether a provider trains on prompts is a declared value from our terms with them. Retention is determined from the upstream listing for every host this model can reach. Neither is guessed.

Measured performance

Time to first response (p50)1.7s

Measured through FastMetal's own gateway over the last 30 days (661 first-response samples).

Time to first response is measured to the first streamed chunk; throughput counts all output tokens, including reasoning tokens. These are observations, not a performance guarantee.

Code Examples

curl https://api.fastmetal.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "hy3",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

How Hy3 actually answers

Real responses to our standard prompts, recorded on FastMetal.

Count the number of 'r's in 'strawberry'

Count the number of 'r's in 'strawberry'. Explain your reasoning step by step.

To find the number of times the letter **'r'** appears in the word **"strawberry"**, let's break it down step by step: **Step 1: Write out the word letter by letter.** The word "strawberry" is spelled: **s – t – r – a – w – b – e – r – r – y** **Step 2: Check each letter individually.** 1. **s** – not an 'r' 2. **t** – not an 'r' 3. **r** – this is an **'r'** (1st occurrence) 4. **a** – not an 'r' 5. **w** – not an 'r' 6. **b** – not an 'r' 7. **e** – not an 'r' 8. **r** – this is an **'r'** (2nd occurrence) 9. **r** – this is an **'r'** (3rd occurrence) 10. **y** – not an 'r' **Step 3: Count the occurrences.** We identified an 'r' at the 3rd, 8th, and 9th positions. 1 + 1 + 1 = **3** *…

11.4s response914 output tokens

Debug This Error

I'm getting the following error in my Node.js application: TypeError: Cannot read properties of undefined (reading 'map') at UserList (/app/components/UserList.js:12:25) at renderWithHooks (/app/node_modules/rea…

### What is causing the error? The error `TypeError: Cannot read properties of undefined (reading 'map')` means that at the moment your `UserList` component is rendered, the `users` prop is **`undefined`**. In JavaScript, you cannot call array methods (like `.map()`) on `undefined`. This usually happens for one of two reasons: 1. **The parent component is not passing the `users` prop** (or is misspelling it, e.g., passing `userList` instead of `users`). 2. **You are fetching data asynchronously** (like from an API), and the initial state is `undefined` before the data arrives. --- ### How to fix it Here are a few ways to resolve this, ranging from quick defensive fixes to best practice…

26.7s response2662 output tokens

Code Review

Please review the following Python function and suggest improvements for readability, performance, and best practices: def get_data(url, retries=3): import requests import time for i in range(retries):…

Here is a review of your function with suggestions categorized by **readability**, **performance**, and **best practices**, followed by a refactored version. --- ## Key Issues in the Original Code ### 1. Readability - **Imports inside the function:** `import requests` and `import time` are executed every time the function is called, which hides dependencies and clutters the logic. - **Unpythonic boolean check:** `if item["active"] == True:` is redundant; Pythonic code simply uses `if item["active"]:`. - **Loop with `.append()`:** A list comprehension is more concise and easier to read. - **No documentation or type hints:** It’s unclear what the function returns or what types are expected.…

61.4s response4082 output tokens

Compare these answers side by side with other models →

Frequently asked questions

How much does the Hy3 API cost?: On FastMetal, Hy3 is billed per token: $0.1477 per 1M input tokens and $0.6119 per 1M output tokens on a US-dollar account, before tax. Usage is drawn from a prepaid balance; there is no subscription or monthly fee.
Can I call Hy3 with the OpenAI SDK?: Yes. Point base_url at https://api.fastmetal.ai/v1 and pass "hy3" as the model; existing OpenAI-style code works unchanged, including streaming, tool calls and structured output.
What is the context window of Hy3?: 262,144 tokens, shared between the prompt and the response.
How does Hy3 rank?: #79 on the public arena's Japanese board (ELO 1,413). Ranks move as the leaderboard is updated.
What do I need to try Hy3?: Create an account and add credit; the model is then available both in the browser chat and over the API. There is no contract or minimum spend.

Try Hy3 right now

Hy3 is available on FastMetal through one API key. Start in the browser, or call it from the OpenAI SDK.

Create a free account

Leaderboard

Text

russianELO: 1,452

#71

CodingELO: 1,503

#73

MathELO: 1,484

#36

Creative WritingELO: 1,424

#78

Instruction FollowingELO: 1,448

#68

Hard PromptsELO: 1,476

#72

Multi-TurnELO: 1,453

#83

OverallELO: 1,454

#74

JapaneseELO: 1,413

#79

ChineseELO: 1,500

#57

KoreanELO: 1,413

#72

EnglishELO: 1,464

#71

frenchELO: 1,473

#74

germanELO: 1,462

#55

spanishELO: 1,471

#31

Code

OverallELO: 1,506

#62

Rankings: LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), CC BY 4.0, snapshot 2026-10-22; category names mapped and ratings rounded by FastMetal.

More from Tencent