// Tencent Hunyuan / Hy · Text LLM
Hunyuan-7B-Instruct: VRAM & local deployment
7.5B parameters. About 6.6 GB at 4-bit and 18.1 GB at BF16 with one 8K-token request.
released 2025-07-30licence: Licence not statedmemory figures are estimates
01Specs
| Publisher | Tencent · tencent |
|---|---|
| Family / series | Tencent Hunyuan / Hy · Hunyuan |
| Type | Text LLM (text-generation) |
| Parameters | 7.5B Source: counted from official safetensors weights |
| Context length | 32K tokens (32,768)max_position_embeddings in config.json |
| Architecture | 32 layers · GQA |
| Released weights | BF16 · 15.0 GBdownload |
| Licence | · Licence not statedThe model card metadata does not state a licence; check the repository files. |
| Released | 2025-07-30 (Hugging Face repo created) |
02How much VRAM it needs (estimate)
Total = weights + 10% + 0.5GB overhead + KV cache for one request (FP16 KV). More concurrent users multiply the KV column.
| Precision | Weights | Total @ 8K | Total @ 32K | Smallest GPU | Mac |
|---|---|---|---|---|---|
| BF16 / FP16 (original) | 15.0 GB | 18.1 GB | 21.3 GB | RTX 4090 24GB | Mac 36GB |
| FP8 / INT8 / Q8_0 | 8.0 GB | 10.3 GB | 13.6 GB | 12GB GPU | Mac 24GB |
| 4-bit (Q4_K_M / AWQ / GPTQ-Int4) | 4.5 GB | 6.6 GB | 9.8 GB | 8GB GPU | Mac 16GB |
KV cache per request: 8K 1.1 GB · 32K 4.3 GB
Bits per weight: BF16 16, FP8 8.5, 4-bit 4.85. “Fits” = total ≤ 90% of memory; Mac GPU can use ~2/3 of unified memory (≤36GB) or ~3/4 (larger). For MoE, all experts must be in memory, so size by total parameters. Method →
03How to run it
Only runtimes the publisher or the runtime’s official library names for this model.
- vLLM — mentioned in the official model card
- SGLang — mentioned in the official model card
- Transformers — mentioned in the official model card
- TensorRT-LLM — mentioned in the official model card
04Official repositories & quantised builds
4 official repo(s) from tencent.
- Hunyuan-7B-Instruct Instruct
- Hunyuan-7B-Instruct-AWQ-Int4 Instruct/AWQ/INT4
- Hunyuan-7B-Instruct-GPTQ-Int4 Instruct/GPTQ/INT4
- Hunyuan-7B-Instruct-FP8 Instruct/FP8
05Other Tencent Hunyuan / Hy models
All 19 →| Model | Params | 4-bit total | Released |
|---|---|---|---|
| Hunyuan-0.5B-Instruct · same series | 539M | 1.7 GB | 2025-07-30 |
| Hunyuan-1.8B-Instruct · same series | 1.79B | 2.2 GB | 2025-07-30 |
| Hunyuan-4B-Instruct · same series | 4.22B | 4.5 GB | 2025-07-30 |
| Hunyuan-A13B-Instruct · same series | 80.4B / 13B | 55.2 GB | 2025-06-25 |
| Hy-Memory-30B-A3B | 30.1B / 3B | 21.4 GB | 2026-10-15 |
| Hy4-preview | 770B / 49B | 515 GB | 2026-08-27 |
| Hy3 | 295B / 21B | 200 GB | 2026-07-02 |
| Hy-MT2-1.8B | 2.04B | 2.4 GB | 2026-05-11 |
| Hy-MT2-7B | 8.03B | 6.9 GB | 2026-05-11 |
| Hy-MT2-30B-A3B | 30.1B / 3B | 21.4 GB | 2026-05-11 |
| Hy3-preview | 295B / 21B | 200 GB | 2026-04-13 |
| HY-MT1.5-1.8B | 2.04B | 2.4 GB | 2025-12-25 |
06Sources
Data checked 2025-09-02. Parameters, context and licence are copied from the official repository; memory figures are our estimates.