AIHubMixAIHubMix

Hunyuan Models

5 modelsGeneral models from $0.16/M inputUp to 256K context

Usage

Last 30 days · 2026-07-13 to 2026-08-11

Tokens

1.6B

Requests

17.5K

Models in use

5 of 5

Tokens per day

0112M223M07-1307-2007-2708-0308-102026-07-13 — 12,354,500 tokens2026-07-14 — 17,568,955 tokens2026-07-15 — 2,433,490 tokens2026-07-16 — 6,764,400 tokens2026-07-17 — 68,666,615 tokens2026-07-18 — 15,459,410 tokens2026-07-19 — 17,933,820 tokens2026-07-20 — 30,439,185 tokens2026-07-21 — 48,466,205 tokens2026-07-22 — 26,194,360 tokens2026-07-23 — 61,426,370 tokens2026-07-24 — 223,200,765 tokens2026-07-25 — 34,724,620 tokens2026-07-26 — 11,232,375 tokens2026-07-27 — 63,144,575 tokens2026-07-28 — 213,100,060 tokens2026-07-29 — 120,068,380 tokens2026-07-30 — 96,116,450 tokens2026-07-31 — 71,883,405 tokens2026-08-01 — 11,778,695 tokens2026-08-02 — 12,660,305 tokens2026-08-03 — 22,430,265 tokens2026-08-04 — 116,711,225 tokens2026-08-05 — 63,075,145 tokens2026-08-06 — 33,870,465 tokens2026-08-07 — 59,303,490 tokens2026-08-08 — 9,868,600 tokens2026-08-09 — 4,611,335 tokens2026-08-10 — 66,856,290 tokens2026-08-11 — 62,112,215 tokens

Which models that traffic went to

  1. hy389.8%1.4B
  2. hy3-preview8.5%136M
  3. hy-3d-3.11.7%27.9M
  4. tencent/Hunyuan-MT-7B<0.1%210K
  5. tencent/Hunyuan-A13B-Instruct<0.1%39.1K

Share of 1.6B tokens.

The two views disagree on purpose: a model can take a large share of the calls and a small share of the tokens — many short requests — or the reverse. Which one matters depends on whether your cost is driven by call volume or by prompt length. Measured on AIHubMix over the last 30 days, counting the 5 model IDs listed on this page; traffic routed through upstream-specific IDs that are not in the public catalog is not included.

All 5 Hunyuan Models

Open in model list
Hunyuan models on AIHubMix with input and output modalities, context length, maximum output, price per million tokens including cache read and cache write rates, and measured throughput and latency.
Modalities
hy3Takes text, returns text.256K128K$0.16$0.62/M$0.04/M33 tok/s3.34 s
hy3-previewTakes text, returns text.256K128K$0.17$0.57/M$0.05/M49 tok/s2.54 s
hy-3d-3.1Takes text, vision. Output modality not published.——FreeFree/M———
tencent/Hunyuan-A13B-Instruct——$0.14$0.56/M———
tencent/Hunyuan-MT-7B——$0.20$0.20/M———

Prices are USD per million tokens; cache read and cache write are the rates for prompt-cache hits and for writing a prompt into the cache. Throughput and latency are measured on AIHubMix — the same figures the model detail page shows — not vendor claims. A dash means the catalog does not publish that field for that model, which is not the same as the model not supporting it.

Common Comparisons

hy3 vs hy3-previewhy3 vs qwen3.8-max-previewhy3 vs kimi-k3glm-5.2 vs hy3grok-4.5 vs hy3

Hunyuan on AIHubMix

Which Hunyuan model should I start with?

hy3 at $0.16/M input — the cheapest entry here that declares tool calling, and it carries a 256K context. Move up to tencent/Hunyuan-MT-7B when answer quality matters more than cost, or to hy3-preview for long-form reasoning.

Which of these models reason before answering?

2 of the 5 models here declare a reasoning phase — they work through the problem before producing an answer, which helps on multi-step problems at the cost of extra output tokens. Use the Reasoning filter above the table to see them. The catalog does not record anything further about how they differ, so this page does not sort them into families.

Why are there several entries for the same model?

Because each row is a route you can call, not a model release. Some IDs name an upstream (azure-, alicloud-, cc-), some are the open-weight repository form (tencent/…), and some differ only in capitalisation, kept so older integrations keep working.

The catalog does not carry a field saying which of those a given row is, so this page does not sort them into buckets it would have to invent. Every row shows that route’s own price, context and speed — compare those directly, and open a model to see the upstreams that serve it.

How is cached input billed?

The Cache read column is the rate for input tokens served from the prompt cache — for example hy3 bills cache hits at 25% of the input rate and hy3-preview bills cache hits at 30% of the input rate. Cache write is the surcharge for putting a prompt into the cache in the first place, and only a few upstreams bill it separately. A dash in either column means the catalog carries no cache rate for that model, so plan on paying the full input rate.

Do I need a separate Hunyuan account?

No. One AIHubMix key covers every model on this page, and switching between them is a change to the model string — billing, rate limits, and logs stay in one place.

Other Providers

Qwen
140 modelsOpenAI
132 modelsGoogle
81 modelsZ.AI
64 modelsLlama
52 models

Start calling Hunyuan in one line

One key, one endpoint, 844 models across 35 providers.

Get API keyRead the docs
AIHubMix

© 2023 - 2026 AIHubMix, LLC

Product

Company

Developer

Connect