Tencent › Hy4 preview › Pricing

Hy4 preview pricing

API list price by Tencent · same on DeepInfra, Novita, SiliconFlow · checked Oct 23, 2026
Benchmarks and alternatives →
$0.83 / $2.50per 1M tokens in / out
$0.042cache read · −95%
#6on OpenRouter by tokens · 8.5T / week

Price sheetper 1M tokens

Input$0.83list
Output$2.50
Cache read$0.042−95% vs input
Context · max output1M · 64Ktokens

Where to buy4 providers · per 1M tokens

ProviderInputOutputContext
DeepInfra$0.83$2.501M
Novita$0.83$2.501M
SiliconFlow$0.83$2.501M
Tencentlab$0.83$2.501M

In Tencent's line-upper 1M in / out

Hy MT2 1.8B$0.044 / $0.18−94%
Hy MT2 30B A3B$0.074 / $0.29−90%
Hy MT2 7B$0.074 / $0.29−90%
Hy3$0.13 / $0.53−82%
Hunyuan A13B Instruct$0.14 / $0.57−80%
Hy4 preview$0.83 / $2.50this model
<

Cheaper for the same workvs Hy4 preview

DeepSeek V4.1 Flash 1017 maxOpen-58%score wins 55%$0.30 / $1.20
MiMo v2.6 Flash unstatedOpen-86%same score$0.14 / $0.28
Qwen3.8 Flash unstatedOpen-82%same score$0.15 / $0.47
MiMo v2.6 Pro unstatedOpen-57%same score$0.43 / $0.87
GLM 5.3 Flash unstatedOpen-81%score wins 47%$0.15 / $0.50

Pricing questions

How much does Hy4 preview cost?›

$0.83 per 1M input tokens and $2.50 per 1M output tokens on Tencent's API and DeepInfra, Novita, SiliconFlow. Cached input is $0.042.

Is Hy4 preview free to use?›

There is no free API tier, but the weights are open: you can run it on your own hardware for the cost of compute.

How much does prompt caching save?›

Cache reads cost $0.042 per 1M, 95% below the input price.

What is a cheaper alternative to Hy4 preview?›

DeepSeek V4.1 Flash 1017 at max effort costs -58% per task with a score wins 55% on Vals LegalBench.