# hy4-flash — API, Pricing & Context Window | Vivgrid

> Source: https://vivgrid.com/docs/models/hy4-flash

hy4-flash on Vivgrid: Tencent's low-cost Hy4 model with image input, a 1M-token context window, and 64K output.

`hy4-flash` is the lower-cost tier of Tencent's Hy4 line, built for everyday agent and productivity tasks. It reads text and images, supports function calling and deep reasoning, and pairs a **1M-token context window** with **64K-token max output** at $0.07 per 1M input tokens and $0.281 per 1M output tokens.

On Vivgrid, `hy4-flash` is served on **Chat Completions** (`/chat/completions`) through the unified, OpenAI-compatible endpoint, so it sits alongside models from OpenAI, Anthropic, Google, and DeepSeek under one API key.

## Specifications

| Field | Value |
| --- | --- |
| Provider | Tencent |
| Model ID | `hy4-flash` |
| Best for | General-purpose |
| Available on | Chat Completions |
| Context window | 1,048,576 tokens |
| Max output | 65,536 tokens |
| Modalities | Text, Image |
| Tool / function calling | Yes |
| Knowledge cutoff | — |
| Acceleration | Global (centralized) |

## Pricing

Pricing in USD per 1M tokens.

| Input | Cached input | Output |
| --- | --- | --- |
| $0.07 | $0.01 | $0.281 |

### What one request costs

The rates above applied to a 20K-token prompt with a 2K-token reply — a typical single agent turn, small enough that every request is billed at the base rate.

| Request | Cost |
| --- | --- |
| Cold prompt, nothing cached | $0.00196 |
| Warm prompt, 90% of the input served from cache | $0.00088 |


## API surfaces

The Vivgrid Console catalogs `hy4-flash` on one surface, all behind the same API key. The endpoint shown is the one the Console previews for that surface.

| Surface | Primary endpoint | Use it for |
| --- | --- | --- |
| Chat Completions | `/chat/completions` | Agent projects, where Vivgrid injects the model, system prompt and tools server-side, and Model API calls that name the model per request. |

## Quick start

Get an API key from the Vivgrid Console (https://console.vivgrid.com), then call `hy4-flash` directly.

```bash
curl https://api.vivgrid.com/v1/chat/completions \
  -H "Authorization: Bearer $VIVGRID_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "hy4-flash",
    "messages": [
      { "role": "user", "content": "Say hello in English, Chinese and Spanish." }
    ],
    "stream": true
  }'
```


## Ideal use cases

- High-volume, cost-sensitive chat and agent traffic that still needs tool calling
- Screenshot, document, and chart understanding at low cost
- Long-document reasoning within a 1M-token context
- Classification, extraction, and routing steps ahead of a larger model

## Related models

- [glm-5.3-flash](/docs/models/glm-5.3-flash) — Zhipu AI's low-cost multimodal model with image, video, and audio input
- [deepseek-v4.1-flash](/docs/models/deepseek-v4.1-flash) — DeepSeek's fast, low-cost model with image input and 384K output
- [gpt-6-luna](/docs/models/gpt-6-luna) — OpenAI's fast, low-cost gpt-6 model
