One agent session
1M input tokens and 100K output tokens, no caching.
- Input1M x $0.834 = $0.834
- Output0.1M x $2.501 = $0.2501
$1.08per session
Tencent - Open weights
Hy4 preview is an open-weights model from Tencent that costs $0.834 per 1M input tokens and $2.501 per 1M output tokens on the first-party API, with a 1M-token context window.
Last verified from tencentcloud.com. All sources
| Item | Price / 1M tokens |
|---|---|
| Input | $0.834 |
| Output | $2.501 |
| Cache read | $0.042 |
Note: Sold through Tencent Cloud TokenHub (international), Tencent's first-party LLM service. The pricing page has a single 'Cache' column (USD per 1M tokens), recorded as cache read; the launch post calls it the price 'for cache hits'. No separate cache-write, batch or peak/off-peak price is listed for Hy models. The price is the same in both regional language-model tables on the pricing page (Singapore/Guangzhou tabs). The model list gives context as '1M' with max input 960k and max output 64k; the launch post says the context window exceeds 1M tokens. 770B total / 49B active parameters. Release date from the Tencent launch post (published 2026-08-28). Vercel AI Gateway lists it with a 1,024,000 context; the first-party figure is recorded here. It is a preview model.
hy4-previewtencent/hy4-previewTokenHub model list shows Deep Reasoning (Preserved Thinking), Structured Output, Function Calling and Caching. Not stated means the provider's docs do not say either way. It is not the same as No. Checked against provider docs on . Compare capabilities across models.
Worked from the standard list rates above, with the math shown so you can swap in your own token counts.
1M input tokens and 100K output tokens, no caching.
$1.08per session
10,000 requests a day at 2K input and 500 output tokens each.
$29.19per day
$875.55 per 30-day month
90% of the input is read from the prompt cache. The one-time cache write premium is not included.
$0.3713per session
| Model | Input / 1M | Output / 1M |
|---|---|---|
| Hy4 preview (this page) | $0.834 | $2.501 |
| Amazon Nova 2 LiteAmazon | $0.30 | $2.50 |
| Dola Seed 2.1 TurboByteDance Seed | $0.50 | $2.50 |
| Gemini 2.5 FlashGoogle | $0.30 | $2.50 |
| Gemini 3.5 Flash-LiteGoogle | $0.30 | $2.50 |
| Hy3Tencent | $0.132 | $0.528 |
| Hy-MT2-PlusTencent | $0.074 | $0.295 |
| Hy-MT2-ProTencent | $0.074 | $0.295 |
| Hy-MT2-LiteTencent | $0.044 | $0.177 |
Hy4 preview costs $0.834 per 1M input tokens and $2.501 per 1M output tokens on Tencent's first-party API, per tencentcloud.com on October 17, 2026. Per 1M tokens, cache reads cost $0.042.
Hy4 preview has a 1,000,000-token context window and returns up to 64,000 output tokens per response.
Yes. Hy4 preview's weights are published at https://huggingface.co/tencent/Hy4-preview under the Apache-2.0.
Tencent released Hy4 preview on August 28, 2026.
Last verified from tencentcloud.com.
Newsletter
One email per week. New model releases, price cuts and what they mean for your bill.