Hy4 preview
by TencentChat
A next-generation productivity model with significantly enhanced Agent and complex task execution capabilities.
- Context
- 1.05M tokens
- Max output
- 65,536 tokens
- Parameters
- 780B total
- Released
- 2026-08-28
- Weights
- Open
- Licence
- apache-2.0
tencent/hy4-preview · also seen as deepinfra/tencent/Hy4-preview, tencent/hy4-preview-20260827
Specifications
Full facts as recorded, incl. what is unknown.
- Released
- 2026-08-28
- Weights
- Opennot gated
- License
- apache-2.0
- Input
- Text
- Output
- Text
- Context window
- 1,048,576 max across endpoints
- Max output
- 65,536
- Providers
- 10 + 11 gateways, not counted as providers
- Epoch Capabilities Index
- not evaluated
- Direct price range
- $0.67–$1.15 in · $2.00–$3.44 out per 1M tokens · 9 of 10 providers have a comparable price
More details
- Last updated
- 2026-10-15
- Knowledge cutoff
- Weights link
- Hugging Face · tencent/Hy4-preview
- Parameters
- 780B total
- Developer country
- China
Capabilities
Across all endpoints: supported if any endpoint supports it.
- Tool calling
- Parallel toolsunknown
- Structured output
- Reasoning
- Image input
- PDF input
- Audio input
- Audio outputunknown
- Video input
- Prompt caching
- Web search
- Computer useunknown
- System messagesunknown
- Assistant prefillunknown
- Temperature
- Streamingunknown
- Attachments
Endpoints
One row per provider and variant; gateways are listed separately.
Loading endpoints…
Price history
Atom feedNo change since 19 Oct 2026. The cheapest direct price has been $0.67 in / $2.00 out per 1M tokens at Vancine on every data refresh since then.
Tracked from 19 Oct 2026. Headline-eligible direct prices only, so a price the sanity check flags never draws a spike. Series as JSON
Cost calculator
Loading endpoints…
Similarly priced models
Blended rate per 1M tokens, each at its own cheapest provider (10M input : 2M output mix).
| Model | Blended/1M | In/1M | Out/1M | Context | vs Hy4 preview |
|---|---|---|---|---|---|
| GLM-5V-TurboZ.ai (Zhipu AI) | $1.10 | $0.704 | $3.10 | 200K | −1% cheaper |
| Qwen3.7 MaxAlibaba (Qwen) | $1.10 | $0.825 | $2.48 | 1M | −1% cheaper |
| Morph v3 LargeMorph | $1.07 | $0.90 | $1.90 | 262K | −4% cheaper |
| GPT-3.5 TurboOpenAI | $1.17 | $1.00 | $2.00 | 16.4K | +5% |
| Grok 3 Mini FastxAI | $1.17 | $0.60 | $4.00 | 131K | +5% |
| Grok Build 0.1xAI | $1.17 | $1.00 | $2.00 | 256K | +5% |
| Apertus 70B InstructSwiss AI | $1.17 | $0.82 | $2.92 | 65.5K | +5% |
| Seed 2.1 ProByteDance Seed | $1.18 | $0.707 | $3.54 | 256K | +6% |
More from Tencent
Newest Tencent models in the catalogue. All Tencent models
| Model | Released | Context | Listed price |
|---|---|---|---|
| Hy Image 3.5 Preview | 2026-10-05 | 100K | unpriced |
| Hy-MT2-1.8B | 2026-08-20 | 8.2K | from $0.044 |
| Hy-MT2-30B-A3B | 2026-08-20 | 8.2K | from $0.074 |
| Hy-MT2-7B | 2026-08-19 | 8.2K | from $0.074 |
| Hy3 | 2026-07-06 | 262K | from $0.132 |
| Tencent Hy-MT2-Lite | 2026-06-12 | 8K | $0.044 via Vercel AI Gateway |
| Tencent Hy-MT2-Pro | 2026-05-21 | 8K | $0.074 via Vercel AI Gateway |
| Tencent Hy-MT2-Plus | 2026-05-21 | 8.2K | $0.074 via DevPass (LLM Gateway) |