This model is deprecated. We still benchmark its performance, but less often: every 18 hours instead of every 6.
Tencent has launched a newer release, Hy3. We suggest considering it instead.
Hy3-preview: Intelligence, Performance & Price Analysis
Available providers
Model summary
Performance
Verbosity
Hy3-preview (Reasoning) supports text input, outputs text, and has a 256k tokens context window.
Pricing for Hy3-preview (Reasoning) is $0.063 per 1M input tokens (competitively priced, median: $0.43) and $0.21 per 1M output tokens (competitively priced, median: $1.84).
| Input modality | Supports: text |
|---|---|
| Output modality | Supports: text |
| Context window | 256k ~384 A4 pages of size 12 Arial font |
| Total parameters | 295B |
| Active parameters | 21B Number of parameters active per token during inference |
| License | TENCENT HY COMMUNITY LICENSE AGREEMENT |
| Model weights | Hugging Face |
Metrics are compared against models of the same class:
- Non-reasoning models → compared only with other non-reasoning models
- Reasoning models → compared across both reasoning and non-reasoning
- Open weights models → compared only with other open weights models of the same size class:
- Tiny: ≤4B parameters
- Small: 4B–40B parameters
- Medium: 40B–150B parameters
- Large: >150B parameters
- Proprietary models → compared across proprietary and open weights models of the same price range, using a blended 3:1 input/output price ratio:
- <$0.15 per 1M tokens
- $0.15–$1 per 1M tokens
- >$1 per 1M tokens
Time per Task
Intelligence
Artificial Analysis Intelligence Index
Artificial Analysis Intelligence Index by Open Weights / Proprietary
Intelligence Index vs. Cost per Task
Openness Index
Artificial Analysis Openness Index: Score
Cost
Cost per Intelligence Index Task
Cost to Run Artificial Analysis Intelligence Index
Pricing: Cache Hit, Input, and Output
Context Window
Context Window
Model Size (Open Weights Models Only)
Model Size: Total and Active Parameters
Further details
More from Tencent
Frequently Asked Questions
Common questions about Hy3-preview
Hy3-preview was released on April 23, 2026.
Hy3-preview was created by Tencent.
Hy3-preview costs $0.063 per 1M input tokens (very competitive, median: $0.43) and $0.21 per 1M output tokens (very competitive, median: $1.84), based on the median across providers serving the model.
Hy3-preview costs $0.063 per 1M input tokens and $0.21 per 1M output tokens (based on the median across providers serving the model). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.0483 per 1M tokens. Pricing may vary by provider. Compare provider pricing
Yes, Hy3-preview (Reasoning) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer. No, Hy3-preview (Non-reasoning) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Hy3-preview supports text input.
Hy3-preview supports text output.
No, Hy3-preview does not support image input. It can only process text.
No, Hy3-preview is not multimodal. It only supports text input.
Hy3-preview has a context window of 256k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Hy3-preview is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Hy3-preview has 295B parameters (21B active).
Hy3-preview is a Mixture of Experts (MoE) model with 295B total parameters, but only 21B active parameters are used during inference.
Hy3-preview is released under the TENCENT HY COMMUNITY LICENSE AGREEMENT license. Commercial use is allowed with restrictions. View license
Yes, Hy3-preview is available via API through 1 provider. Compare API providers
Hy3-preview is available through 1 API provider. Compare providers