Check my GPU

LLMs ·

Hunyuan A13B (80B-A13B)

Parameters
80B (13.0B active)
Architecture
Mixture of experts
Max context
32K tokens
License
tencent-hunyuan-a13b
Released
2025-06

How to run it

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

huggingface.co/tencent/Hunyuan-A13B-Instruct-GGUF (opens in a new tab)

ollama run hf.co/tencent/Hunyuan-A13B-Instruct-GGUF

Original weights · Hugging Face

Full-precision files from Tencent, for fine-tuning, vLLM, Transformers.

huggingface.co/tencent/Hunyuan-A13B-Instruct (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from Tencent. The weights themselves download from Hugging Face.

github.com/Tencent-Hunyuan/Hunyuan-A13B (opens in a new tab)

git clone https://github.com/Tencent-Hunyuan/Hunyuan-A13B

Links checked 5 days ago against Hugging Face, Ollama and GitHub.

Memory needed

Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps.

Quantization Quality Weights Total at 8K context Total at 32K context Download
Q2_K Low 32.2 GB 33.8 GB 37.1 GB GGUF files (opens in a new tab)
Q3_K_M Low 39.4 GB 41.1 GB 44.3 GB GGUF files (opens in a new tab)
Q4_K_M Good 49.0 GB 50.7 GB 53.9 GB GGUF (opens in a new tab)
Q5_K_M High 57.1 GB 58.8 GB 62.0 GB GGUF files (opens in a new tab)
Q6_K High 65.9 GB 67.6 GB 70.8 GB GGUF files (opens in a new tab)
Q8_0 Very high 85.2 GB 86.9 GB 90.1 GB GGUF · part 1 of 2 (opens in a new tab)
F16 Full 161 GB 162 GB 166 GB Original weights (opens in a new tab)

Best hardware for Hunyuan A13B (80B-A13B)

The least expensive GPU in our catalog that runs it without offloading is the AMD Radeon PRO W7800 48GB (launch price $2,799) at Q3_K_M.

Grade · points Hardware Quant Memory Speed Fit
87 ptsExcellent (S) · 87 of 100 points: Model quality 33.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 NVIDIA RTX PRO 6000 Blackwell Q6_K 72.4 GB / 96.0 GB 58.7 tok/s Runs well
81 ptsGreat (A) · 81 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 11.2/15, Memory use 14.0/15 Apple M4 Max (128 GB) Q4_K_M 58.4 GB / 96.0 GB 22.2 tok/s Runs well
80 ptsGreat (A) · 80 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 13.5/15, Memory use 10.0/15 Apple M2 Ultra (192 GB) Q4_K_M 62.2 GB / 144 GB 32.5 tok/s Runs well
79 ptsGreat (A) · 79 of 100 points: Model quality 28.0/45, Memory fit 24.0/25, Speed 12.7/15, Memory use 15.0/15 Apple M4 Max (64 GB) Q3_K_M 44.9 GB / 48.0 GB 28.7 tok/s Tight fit
79 ptsGreat (A) · 79 of 100 points: Model quality 33.0/45, Memory fit 25.0/25, Speed 6.4/15, Memory use 15.0/15 AMD Ryzen AI Max+ 395 (128 GB) Q6_K 75.3 GB / 96.0 GB 7.2 tok/s Runs well
78 ptsGreat (A) · 78 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 13.7/15, Memory use 7.9/15 Apple M3 Ultra (256 GB) Q4_K_M 66.1 GB / 192 GB 33.3 tok/s Runs well
77 ptsGreat (A) · 77 of 100 points: Model quality 28.0/45, Memory fit 24.0/25, Speed 10.9/15, Memory use 15.0/15 Apple M1 Max (64 GB) Q3_K_M 44.9 GB / 48.0 GB 21.0 tok/s Tight fit
77 ptsGreat (A) · 77 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 8.5/15, Memory use 12.0/15 NVIDIA GB10 (128 GB) Q4_K_M 59.9 GB / 115 GB 12.9 tok/s Runs well
75 ptsGreat (A) · 75 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 13.7/15, Memory use 4.9/15 Apple M3 Ultra (512 GB) Q4_K_M 81.4 GB / 384 GB 33.3 tok/s Runs well
62 ptsGood (B) · 62 of 100 points: Model quality 23.3/45, Memory fit 10.0/25, Speed 14.4/15, Memory use 15.0/15 NVIDIA GeForce RTX 5090 Q2_K 35.4 GB / 32.0 GB 37.0 tok/s Runs with offload

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.