Hy4-preview
Hy4-preview is an open-weight language model from Tencent Hunyuan with 780B parameters (mixture of experts). At 4-bit it needs about 516.2 GB of memory, more than any single device in our directory has.
Overview
Includes room for an 8k context and runtime overhead. More context needs more memory.
Key information
- Lab
- Tencent Hunyuan
- Released
- 27 Aug 2026
- Parameters
- 780B · MoE
- Type
- Language
- Context
- 1,048,576 tokens
- License
- Apache-2.0
- HF downloads, 30 days
- 11,554
- Repository
- tencent/Hy4-preview
More from Tencent Hunyuan
View allWill it run on your hardware?
How we estimate| Estimate or device | Memory |
|---|---|
| Q4 | 516.2 GB |
| Q8 | 910.9 GB |
| BF16 | 1717.4 GB |
| RTX 5060 Ti 16 GB | 16 GB usable |
| RTX 5090 32 GB | 32 GB usable |
| DGX Spark 128 GB | 108.8 GB usable |
| Mac Studio M5 Ultra 256 GB | 192 GB usable |
Single-device memory estimate. Pick a device to update the On your device column.
36 GB usable memory
| Format | Memory needed | On your device | Example hardware |
|---|---|---|---|
| 4-bit (Q4) | ≈ 516.2 GB | Too big 1434% | No single device in our directory has enough memory |
| 8-bit (Q8) | ≈ 910.9 GB | Too big 2530% | No single device in our directory has enough memory |
| 16-bit (BF16) | ≈ 1717.4 GB | Too big 4771% | No single device in our directory has enough memory |
Fits means the estimate stays within 90% of usable memory. Tight is above 90% and up to 100%. Too big is above 100%. These are memory checks, not speed or compatibility tests.
Mac allowances are 66% of total RAM up to 36 GB, and 75% above. AI boxes allow 85% because their memory is shared with the operating system.
Where to run it
On a Mac
Start with the MLX files, built for Apple Silicon.
mlx-community/Hy4-preview-4bitCheck your MacThrough an API
AtlasCloud (fp8) is the cheapest provider we found.
$0.834 / $2.501
Input / output per 1M tokens
₹79 / ₹236 at ₹94.29 per $1
Blended: $1.2508 (₹118) per 1M. Ranked by a 3:1 mix of input and output tokens. Tied with DeepInfra, GMICloud, Novita, SiliconFlow and Tencent.
mlx_lm.generate --model mlx-community/Hy4-preview-4bit \
--prompt "Explain how a transformer works."Example commands built from the repository ids above. They need a working install of MLX. We have not run them; check each repository for available files.
| Precision | |||||
|---|---|---|---|---|---|
| AtlasCloudCheapest | $1.2508₹118 | $0.834₹79 | $2.501₹236 | 1,048,576 | FP8 |
| DeepInfra | $1.2508₹118 | $0.834₹79 | $2.501₹236 | 1,048,576 | FP4 |
| GMICloud | $1.2508₹118 | $0.834₹79 | $2.501₹236 | 262,144 | BF16 |
| Novita | $1.2508₹118 | $0.834₹79 | $2.501₹236 | 1,000,000 | FP8 |
| SiliconFlow | $1.2508₹118 | $0.834₹79 | $2.501₹236 | 1,048,576 | FP8 |
| Tencent | $1.2508₹118 | $0.834₹79 | $2.501₹236 | 1,048,576 | FP8 |
INR at ₹94.29 per $1. Prices fetched 14 Nov 2026.
More about this model
Frequently asked questions
How much memory does Hy4-preview need? At an 8k context, about 516.2 GB at 4-bit, 910.9 GB at 8-bit and 1717.4 GB at 16-bit (BF16). A longer context needs more.
Can I run Hy4-preview on a Mac? No Mac in our directory has enough usable memory for the 4-bit estimate of 516.2 GB. An MLX build is available: mlx-community/Hy4-preview-4bit.
Does the full context fit? Hy4-preview supports up to 1,048,576 tokens of context. Our memory estimates use 8k, so long prompts need more memory than shown.
What does it cost through an API? The cheapest provider we found is AtlasCloud, at $0.83 per 1M input tokens and $2.50 per 1M output tokens (₹79 and ₹236 at 94.29 rupees per dollar). Ranked by a 3:1 mix of input and output tokens. Tied with DeepInfra, GMICloud, Novita, SiliconFlow and Tencent.
License in plain English
Apache-2.0. This license permits commercial use, modification and redistribution, with conditions such as keeping the required notices. This summary is not legal advice.
OMHQ RigBench
Not yet tested. We show no speed or quality score without a measurement.
Revision history and sources
Last updated . Sources:Hugging Face model card, MLX repository, OpenRouter.