Hy3 API pricing
Tencent · Budget price tier · Apache 2.0 · released 2026-07. Prices in USD per 1 million tokens. Data snapshot: 2026-11-12.
A support ticket workload of 100K requests a month costs $46.20 on Hy3; the next cheaper model in the Budget price tier for this workload, Qwen3.8 Flash, does it for $46.00. The price we quote for Hy3 has changed 10 times since 2026-08-15, most recently by +60.0% on 2026-10-19; it is now the 80th cheapest of 160 models in the Budget price tier. None of its 4 hosts undercuts the price we quote. It holds the FinOps Friendly badge: top-40% Arena quality, top-30% Efficiency, not among the dearest 30%, and no stability flag.
- Input: $0.132 per 1M tokens
- Output: $0.528 per 1M tokens
- Batch API (in/out): Not published
- Prompt-cache read: $0.033 per 1M tokens
- Context window: 262K tokens
- Arena ELO: 1439 as of 2026-11-08
Market position
On blended price ($0.4092 per 1M tokens at 30% input, 70% output) it is cheaper than 50% of models in the Budget price tier: 80th cheapest of 160. Across every price tier it is cheaper than 74% of priced text models in the catalogue: 80th cheapest of 303. Its Arena ELO of 1439 ranks 54th of the 160 models that carry one, tied with 2 others, above 65% of them. With an Arena score of 1439, Hy3 clears OptimNow's Standard working quality threshold: Standard (Arena 1420+), used for meeting summary, invoice processing and call summary. Each threshold is a distance below the best ranked model, which scores 1530 on Arena scores as of 8 Nov 2026.
Licence, size and capabilities
Hy3 is open source: its weights are published under Apache 2.0, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token. Its published size is 295B parameters. Listed capabilities: Text, Reasoning, Agents. Filling its 262K context window once costs $0.0346 in input tokens at list price.
Pricing by host
A host is a company that runs the model and sells it by the token: the model maker, a cloud platform like Amazon Bedrock, Azure or Vertex, or an independent host. One model, several hosts, and often several prices. This model is sold by 4 different hosts. The price we quote elsewhere on this page is marked. We quote one host's rate so that models stay comparable; it is not a recommendation to buy there.
The price we quote is already the cheapest of the 4 hosts: Tencent at $0.132 input / $0.528 output per 1M tokens. It serves a quantized fp8 build, which is not the same product as the full-precision model. The dearest, AtlasCloud, charges 1.5x the cheapest.
| Host | Input $/1M | Output $/1M | Precision |
|---|---|---|---|
| Tencent (model maker, price we quote) | $0.132 | $0.528 | fp8 |
| Novita | $0.14 | $0.58 | not declared |
| GMICloud | $0.14 | $0.58 | bf16 |
| AtlasCloud | $0.20 | $0.80 | fp8 |
List prices as published on 2026-11-12, via OpenRouter. Negotiated rates, committed-use discounts and provisioned capacity are not shown. These hosts do not all serve the model at the same numeric precision, so the cheapest is not a like-for-like substitute for the dearest.
Monthly cost at 100K requests, by business use case
From the after-discounts figure (prompt caching on the cached input share, batch API for async workloads, only where Tencent publishes those rates), which assumes you implement both, to the list price. One figure means no discount is published.
| Use case | Tokens (in/out) | Monthly (after discounts – list) |
|---|---|---|
| Support Ticket | 1,500 / 500 | $37 – $46 |
| Knowledge Q&A | 2,000 / 800 | $55 – $69 |
| Meeting Summary | 10,000 / 1,200 | $185 – $195 |
| Marketing Content | 2,500 / 1,800 | $123 – $128 |
| Coding Task | 3,000 / 2,000 | $130 – $145 |
| Invoice Processing | 1,500 / 600 | $47 – $51 |
| Call Summary | 2,000 / 700 | $61 – $63 |
| Agent Workflow | 6,000 / 3,000 | $196 – $238 |
Price history
Hy3 has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.132 input / $0.528 output per 1M tokens. The price we quote has changed 10 times since then. The latest change, on 2026-10-19, was +60.0% on the blended price (30% input, 70% output), to $0.132 input / $0.528 output. Net of all changes, the blended price is back where it started on 2026-08-15.
| Date | Old price (in / out) | New price (in / out) | Blended change |
|---|---|---|---|
| 2026-10-19 | $0.0825 / $0.33 | $0.132 / $0.528 | +60.0% |
| 2026-10-18 | $0.132 / $0.528 | $0.0825 / $0.33 | -37.5% |
| 2026-10-04 | $0.0825 / $0.33 | $0.132 / $0.528 | +60.0% |
| 2026-10-03 | $0.132 / $0.528 | $0.0825 / $0.33 | -37.5% |
| 2026-10-02 | $0.0825 / $0.33 | $0.132 / $0.528 | +60.0% |
5 earlier changes are not listed.
Alternatives to Hy3
2 of the 4 comparable models below cost less than Hy3: Gemma 4 26B A4B undercuts it by 77%, for 5 Arena points less.
- Next cheaper model in the Budget price tier: Qwen2.5 72B Instruct (Alibaba) at $0.388 blended, 5% less.
- Next more expensive: Hunyuan A13B Instruct (Tencent) at $0.441 blended, 8% more.
The nearest models in the Budget price tier on Arena ELO, each with its blended price and score difference against Hy3.
- Qwen3.8 27B (Alibaba): $0.15 in / $1.875 out per 1M, Arena ELO 1436 (232% more expensive, 3 ELO points lower)
- Qwen3.6 Plus (Alibaba): $0.325 in / $1.95 out per 1M, Arena ELO 1436 (257% more expensive, 3 ELO points lower)
- Gemma 4 31B (Google): $0.20 in / $0.40 out per 1M, Arena ELO 1443 (17% cheaper, 4 ELO points higher)
- Gemma 4 26B A4B (Google): $0.08 in / $0.10 out per 1M, Arena ELO 1434 (77% cheaper, 5 ELO points lower)
- Cheapest within 15 Arena points: Gemma 4 26B A4B (Google), 77% cheaper, 5 ELO points lower, out of 42 models in that band.
- Highest Arena ELO within 20% of this price: Gemma 4 31B (Google), 17% cheaper, 4 ELO points higher, out of 13 models in that price band.
Other models from Tencent
Tencent has 5 other models in the catalogue. Prices per 1M tokens, input then output.
- Hy-MT2-1.8B: $0.044 / $0.177
- Hy-MT2-30B-A3B: $0.074 / $0.295
- Hy-MT2-7B: $0.074 / $0.295
- Hunyuan A13B Instruct: $0.14 / $0.57
- Hy4: $0.417 / $1.2505
All 6 Tencent models and prices
Common questions about Hy3 pricing
- How much does Hy3 cost?
- Hy3 costs $0.132 per 1M input tokens and $0.528 per 1M output tokens. That is the model maker's own rate (Tencent). Prices as listed on 2026-11-12.
- Which host is cheapest for Hy3?
- The price we quote is already the cheapest of the 4 hosts: Tencent at $0.132 input / $0.528 output per 1M tokens. It serves a quantized fp8 build, which is not the same product as the full-precision model. The dearest, AtlasCloud, charges 1.5x the cheapest. Host rates as listed on 2026-11-12.
- Does Hy3 offer batch or cached pricing?
- Yes, for prompt caching only. Cache reads cost $0.033 per 1M tokens, as listed on 2026-11-12. No batch pricing is published.
- What would Hy3 cost per month?
- With the Support Ticket profile (1,500 input and 500 output tokens per request), 100,000 requests a month cost about $46 on Hy3 at the list price quoted on 2026-11-12.
- Can Hy3 be self-hosted?
- Yes. Hy3 is open source: its weights are published under Apache 2.0, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token.
- Has the price of Hy3 changed?
- Hy3 has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.132 input / $0.528 output per 1M tokens. The price we quote has changed 10 times since then. The latest change, on 2026-10-19, was +60.0% on the blended price (30% input, 70% output), to $0.132 input / $0.528 output. Net of all changes, the blended price is back where it started on 2026-08-15.
Its weights are published, which changes what you may run yourself: the cost gap between open weights and proprietary APIs.
These are list prices. Your real cost also depends on caching and batching, negotiated discounts, marketplace agreements, and the contracts you already have. OptimNow audits exactly that.
Build the business case · Get a FinOps review
LLM pricing · Models by maker · Cloud compute pricing · Compare models · Price barometer · FinOps guides · API and MCP server