RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Moonlight 16B A3B Instruct run on the H200?

Yes. At 8K context the best fit is F16, using about 30.7 GB of the 134 GB usable on the H200. Estimated speed ceiling: ~800 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •7.0 GB Fits7.1 GB Fits3883
3-bit (Q3_K_M) •8.5 GB Fits8.7 GB Fits3082
4-bit (Q4_K_M) •10.6 GB Fits10.8 GB Fits2423
5-bit (Q5_K_M) •12.0 GB Fits12.2 GB Fits2121
6-bit (Q6_K) •14.1 GB Fits14.2 GB Fits1789
8-bit (Q8_0) •16.6 GB Fits16.8 GB Fits1504
16-bit (F16/BF16)30.6 GB Fits30.7 GB Fits800

• real GGUF file size from mmnga/Moonlight-16B-A3B-Instruct-gguf. Speed is a bandwidth ceiling (4800 GB/s), not a benchmark. Methodology.

More

Moonlight 16B A3B 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 (moonshotai/Moonlight-16B-A3B-Instruct), NVIDIA specs. Data as of 2026-11-17.