| Provider | Tencent |
|---|---|
| Model ID | tencent/hy3-preview |
| Prompt Price per 1M tokens | $0.063 |
| Completion Price per 1M tokens | $0.21 |
| Sample Workload Cost 1M input + 500K output | $0.17 |
| Context Window | 262.14K |
| Release Date | |
| Popularity | #2 |
| Popularity Signal | 688.7B tokens |
🔥Hy3 preview
Tencent API model specs with normalized input price, output price, context window, sample workload cost, and related comparisons.
Page updated:
Data confirmed:
Prices normalized to USD per 1M tokens
Sample workload: 1M input + 500K output
Budget brief
Hy3 preview is estimated at $0.17 for the standard workload.
Hy3 preview is best suited for cost-sensitive production traffic. A 1M input token plus 500K output token workload is estimated at $0.17.
Input price$0.063per 1M tokens
Output price$0.21per 1M tokens
Context window262.14Kreported maximum
Popularity#2688.7B tokens signal volume
Use this as a first-pass planning estimate, then verify provider limits, routing, discounts, and availability before production deployment.
Estimate your workload cost
Estimate this model for your workload
This estimate uses normalized public API pricing per 1M tokens. It is a planning aid, not a billing quote. Verify provider pricing, limits, and terms before production use.
Alternative path
Alternative Shortlist
Use these rows to build a test shortlist around Hy3 preview. Lower cost, same-provider fit, and larger context are separate decisions, so each group is ranked by a different signal.
Lower-cost alternatives
Cross-provider candidates with a lower standard 1M input plus 500K output estimate.
| Model | Provider | Sample Cost | Context | Why it is here | Next Step |
|---|---|---|---|---|---|
| 🔥Owl Alpha | OpenRouter | $0 | 1.05M | Lower standard workload estimate from another provider. | Open model Compare |
| New🔥Nemotron 3 Ultra (free) | NVIDIA | $0 | 1M | Lower standard workload estimate from another provider. | Open model Compare |
| 🔥Laguna M.1 (free) | Poolside | $0 | 262.14K | Lower standard workload estimate from another provider. | Open model Compare |
| Nemotron 3 Super (free) | NVIDIA | $0 | 1M | Lower standard workload estimate from another provider. | Open model Compare |
Same-provider swaps
Lower-cost options from the same provider, useful when account setup or procurement is already fixed.
| Model | Provider | Sample Cost | Context | Why it is here | Next Step |
|---|---|---|---|---|---|
| No lower-cost same-provider swap is currently tracked for this model. | |||||
Larger-context upgrades near this budget
Models with more context that stay within a close sample-cost band when price data is available.
| Model | Provider | Sample Cost | Context | Why it is here | Next Step |
|---|---|---|---|---|---|
| Llama 4 Scout | Meta | $0.25 | 10M | More context while staying near this model's sample-cost band. | Open model Compare |
| 🔥Owl Alpha | OpenRouter | $0 | 1.05M | More context while staying near this model's sample-cost band. | Open model Compare |
| 🔥DeepSeek V4 Flash | DeepSeek | $0.18 | 1.05M | More context while staying near this model's sample-cost band. | Open model Compare |
| 🔥MiMo-V2.5 | Xiaomi | $0.24 | 1.05M | More context while staying near this model's sample-cost band. | Open model Compare |
Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use. It supports configurable reasoning levels across disabled, low, and high modes, allowing it to...
Hy3 preview is best suited for cost-sensitive production traffic.
A 1M input token plus 500K output token workload is estimated at $0.17.
Decision Shortcuts
Compare this model
Search head-to-head pages that include Hy3 preview and review input price, output price, context, and sample workload cost.
Find comparisonsTencent catalog
See other Tencent models before narrowing your shortlist.
Open provider hubCheaper alternatives
Start from models sorted by a standard cost estimate when budget is the first constraint.
Browse low-cost modelsLong-context alternatives
Compare large-context models for retrieval, documents, and codebase review.
Browse long-context modelsHigh-Interest Comparisons
Search this model| Comparison | Cost-first Pick | Context Pick |
|---|---|---|
| 🔥Hy3 preview vs 🔥Claude Opus 4.7 | Hy3 preview | Claude Opus 4.7 |
| 🔥DeepSeek V4 Flash vs 🔥Hy3 preview | Hy3 preview | DeepSeek V4 Flash |
| 🔥Hy3 preview vs 🔥DeepSeek V4 Pro | Hy3 preview | DeepSeek V4 Pro |
| 🔥Hy3 preview vs GLM 5.1 | Hy3 preview | Hy3 preview |
| 🔥Hy3 preview vs Kimi K2.6 | Hy3 preview | Tie |
| 🔥Hy3 preview vs Qwen3.6 35B A3B | Hy3 preview | Tie |
| 🔥Hy3 preview vs MiniMax M2.7 | Hy3 preview | Hy3 preview |
| 🔥Hy3 preview vs Nano Banana 2 (Gemini 3.1 Flash Image Preview) | Hy3 preview | Hy3 preview |
Popular Comparisons
Search all comparisons| Comparison | Newest Release |
|---|---|
| 🔥DeepSeek V4 Flash vs 🔥Hy3 preview | |
| 🔥Hy3 preview vs 🔥MiMo-V2.5 | |
| 🔥Hy3 preview vs New🔥MiniMax M3 | |
| 🔥Hy3 preview vs 🔥DeepSeek V4 Pro | |
| 🔥Hy3 preview vs 🔥Claude Sonnet 4.6 | |
| 🔥Hy3 preview vs 🔥Owl Alpha | |
| 🔥Hy3 preview vs 🔥Claude Opus 4.7 | |
| 🔥Hy3 preview vs 🔥DeepSeek V3.2 | |
| 🔥Hy3 preview vs 🔥Gemini 3 Flash Preview | |
| 🔥Hy3 preview vs New🔥Step 3.7 Flash |