Hy3 preview with Jev
Hy3 preview is a model from Tencent, released 2026-04-22. It costs $0.180 per million input tokens and $0.600 per million output tokens, reads up to 262K tokens of context, and writes up to 235,929 tokens in one reply. Through llm11, set model to tencent/hy3-preview.
- Input price
- $0.180 / 1M
- Output price
- $0.600 / 1M
- Context window
- 262K tokens
- Max output
- 236K tokens
- Cached input
- $0.060 / 1M
- Released
- 2026-04-22
- Accepts
- text
- Intelligence index
- 25.3
Read from the provider's prompt cache
Artificial Analysis, via the catalogue
Read from the live catalogue and refreshed daily. At blended list price, 72% of the 332 priced models we can call cost more.
Using Hy3 preview with Jev
Jev is TypeSafe AI's System One model. It reads each request first and decides which model in your pool answers, with a calibrated confidence on that call. Hy3 preview does the answering when Jev sends it work, or when you name it yourself. At list price Hy3 preview falls in the llm11-fast band, but no pack routes to it because it does not advertise structured output, which the verification checks rely on. You can still pin it: pass tencent/hy3-preview as model and routing steps aside while verification still runs.
Jev is new, so we have not published results for Hy3 preview with and without it, and this page does not invent any. Every response comes with a receipt that names the model that answered and what it cost against your most expensive model, so you can measure the pairing on your own traffic.
What a request costs
List price arithmetic, the same sum a receipt does. llm11 adds nothing per request; the only fee is on buying credits.
| Request | Tokens in / out | One | A thousand |
|---|---|---|---|
| Short chat turn | 500 / 200 | $0.00021 | $0.210 |
| Question over retrieved documents | 4,000 / 500 | $0.00102 | $1.02 |
| Long document summary | 30,000 / 1,000 | $0.00600 | $6.00 |
Cheaper models to try
Nearest in price and below Hy3 preview, each within a few points of it on the intelligence index.
| Model | Input / 1M | Output / 1M | Context | Index | Released |
|---|---|---|---|---|---|
| OpenAI GPT-6 Luna | $0.100 | $0.500 | 1.05M | 37.3 | |
| DeepSeek DeepSeek V4 Flash 0423 | $0.076 | $0.153 | 1.05M | 24.2 | |
| DeepSeek DeepSeek V4 Flash 0731 | $0.018 | $0.320 | 1.31M | 34.3 |
Call it
Same OpenAI request shape. Naming the model pins it, so nothing is routed, and the answer is still checked.
python
from openai import OpenAI
client = OpenAI(base_url="https://www.llm11.com/v1", api_key="llm11_live_...")
res = client.chat.completions.create(
model="tencent/hy3-preview",
messages=[{"role": "user", "content": "Hello"}],
)More from Tencent
| Model | Input / 1M | Output / 1M | Context | Released |
|---|---|---|---|---|
| Hy4 preview | $0.834 | $2.50 | 1.05M | |
| Hy-MT2-1.8B | $0.044 | $0.177 | 8K | |
| Hy-MT2-30B-A3B | $0.074 | $0.295 | 8K | |
| Hy-MT2-7B | $0.074 | $0.295 | 8K | |
| Hy3 | $0.132 | $0.528 | 262K | |
| Hunyuan A13B Instruct | $0.140 | $0.570 | 131K |
Questions
- What is Hy3 preview with Jev?
- Jev is TypeSafe AI's System One model. It reads each request first and decides which model in your pool answers, with a calibrated confidence on that call. Hy3 preview does the answering when Jev sends it work, or when you name it yourself. At list price Hy3 preview falls in the llm11-fast band, but no pack routes to it because it does not advertise structured output, which the verification checks rely on. You can still pin it: pass tencent/hy3-preview as model and routing steps aside while verification still runs. Routing decisions are made by Jev, TypeSafe AI's System One model.
- Is Hy3 preview a System One model?
- No. Hy3 preview answers requests. The System One model is Jev, which decides which model answers each one, so the two do different jobs and llm11 uses both.
- How much does Hy3 preview cost?
- $0.180 per million input tokens and $0.600 per million output tokens. A 4,000 token prompt with a 500 token answer costs about $0.00102, so a thousand of them is about $1.02. llm11 passes provider list price through and charges 5% when you buy credits.
- What is the Hy3 preview context window?
- 262,144 tokens, with up to 235,929 tokens in a single reply.
- Does Hy3 preview support tool calling and structured output?
- The catalogue lists tool calling and reasoning controls, but it does not list structured output.
- Can I use Hy3 preview through llm11?
- Yes. Set model to tencent/hy3-preview on the OpenAI-compatible endpoint and the request goes to it directly, with verification still running.
- Does Jev route to Hy3 preview?
- Routing decisions are made by Jev, TypeSafe AI's System One model. At list price Hy3 preview falls in the llm11-fast band, but no pack routes to it because it does not advertise structured output, which the verification checks rely on. You can still pin it: pass tencent/hy3-preview as model and routing steps aside while verification still runs.
- What is a cheaper alternative to Hy3 preview?
- Nearest in price and below it: OpenAI GPT-6 Luna at $0.100 in and $0.500 out, DeepSeek DeepSeek V4 Flash 0423 at $0.076 in and $0.153 out and DeepSeek DeepSeek V4 Flash 0731 at $0.018 in and $0.320 out. Nearest in price and below Hy3 preview, each within a few points of it on the intelligence index.