impulse

Hy3 Preview

Open weightsPreview

Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use.

API id hy3-previewReleased 22 Apr 2026 (Apr 2026)
Input / 1M
$0.18
at GMI Cloud
Output / 1M
$0.60
at GMI Cloud
Context window
262K
tokens
Max output
-
tokens per reply
Hosts
1
1 listings
Released
Apr 2026
22 Apr 2026

Pricing

List prices at GMI Cloud, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.18$0.60
Cached input
67% off input
$0.06

What it costs a month

At GMI Cloud list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$88.20
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$79.20
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$151
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$576
Use your own numbers

Where to run it

1 host serve Hy3 Preview. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
GMI CloudInference
hy3-preview
$0.18 / $0.60

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Open weights
License: other
Runs in about 717.1 GB of GPU memory

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than GMI Cloud.

Smaller Independent models

No cheaper Independent model keeps its key capabilities.

Other frontier labs

  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    30% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    30% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Has every key capability
    8% less

Open-weight picks

  • Mistral Small 2603
    Mistral AI · on Mistral AI
    $0.15 / $0.60
    Has every key capability
    8% less
  • Qwen3.8 27B
    Alibaba Qwen · on EmpirioLabs AI
    $0.17 / $0.50
    Has every key capability
    11% less
  • Nemotron 3.5 Super VL
    NVIDIA · on Kyma API
    $0.081 / $0.472
    Has every key capability
    37% less
  • Gemma 4 31B IT
    Google · on W&B Inference
    $0.10 / $0.34
    Has every key capability
    44% less

Model line

Newer and older versions, at each lab's own price.

  1. Aug 2026Hy4 PreviewLatest$0.417 / $1.25
  2. Apr 2026Hy3 PreviewThis page$0.18 / $0.60

Call it

Using GMI Cloud's API.

Python
from litellm import completion

response = completion(
    model="gmi/hy3-preview",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Hy3 Preview cost?

On GMI Cloud, Hy3 Preview costs $0.18 per million input tokens and $0.60 per million output tokens. Cached input costs $0.06 per million tokens.

Which APIs offer Hy3 Preview?

Hy3 Preview runs on 1 API: GMI Cloud.

What is Hy3 Preview's context window?

262K tokens.

Is Hy3 Preview being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of Hy3 Preview?

Yes. Hy4 Preview is the newest model in this line, released on 28 Aug 2026.

What are cheaper alternatives to Hy3 Preview?

Models that cost less and keep most of what it can do: Claude Haiku 5.5 on Anthropic (30% less), GPT-6 Luna on OpenAI (30% less) and Mistral Small 2603 on Mistral AI (8% less). Savings are on a mix of 3 input tokens to 1 output token.

When was Hy3 Preview released?

Independent released it on 22 Apr 2026.

What is Hy3 Preview's API model id?

Use hy3-preview with GMI Cloud.

Sources

Raw catalog entry for hy3-preview on GMI Cloud
{
  "cache_read_input_token_cost": 6e-8,
  "input_cost_per_token": 1.8e-7,
  "litellm_provider": "gmi",
  "max_input_tokens": 262144,
  "mode": "chat",
  "output_cost_per_token": 6e-7,
  "source": "https://ie-api.gmicloud.ai/v1/artifact/get_public_artifacts"
}