RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Hunyuan A13B Instruct run on the H200?

Yes. At 8K context the best fit is Q8_0, using about 81.2 GB of the 134 GB usable on the H200. Estimated speed ceiling: ~347 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •28.5 GB Fits29.0 GB Fits32.0 GB Fits1009
3-bit (Q3_K_M) •37.1 GB Fits37.6 GB Fits40.6 GB Fits770
4-bit (Q4_K_M) •46.6 GB Fits47.1 GB Fits50.1 GB Fits608
5-bit (Q5_K_M) •54.3 GB Fits54.8 GB Fits57.8 GB Fits520
6-bit (Q6_K)62.5 GB Fits63.0 GB Fits66.0 GB Fits450
8-bit (Q8_0) •80.7 GB Fits81.2 GB Fits84.2 GB Fits347
16-bit (F16/BF16) •151 GB No151 GB No154 GB No—

• real GGUF file size from unsloth/Hunyuan-A13B-Instruct-GGUF. Speed is a bandwidth ceiling (4800 GB/s), not a benchmark. Methodology.

More

Hunyuan A13B Instruct on every GPU and Mac · Every model that runs on the H200 · Other models that fit 64 GB VRAM · Try other settings in the calculator

Data: Hugging Face Hub (tencent/Hunyuan-A13B-Instruct), NVIDIA specs. Data as of 2026-11-17.