Can GOT OCR2 0 run on the H200?
Yes. At 8K context the best fit is F16, using about 2.8 GB of the 134 GB usable on the H200. Estimated speed ceiling: ~3352 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|
| 2-bit (Q2_K) | 1.3 GB Fits | 1.7 GB Fits | 3.9 GB Fits | 17876 |
| 3-bit (Q3_K_M) | 1.4 GB Fits | 1.8 GB Fits | 4.0 GB Fits | 13716 |
| 4-bit (Q4_K_M) | 1.5 GB Fits | 1.8 GB Fits | 4.1 GB Fits | 10967 |
| 5-bit (Q5_K_M) | 1.5 GB Fits | 1.9 GB Fits | 4.2 GB Fits | 9409 |
| 6-bit (Q6_K) | 1.6 GB Fits | 2.0 GB Fits | 4.2 GB Fits | 8175 |
| 8-bit (Q8_0) | 1.8 GB Fits | 2.1 GB Fits | 4.4 GB Fits | 6309 |
| 16-bit (F16/BF16) | 2.4 GB Fits | 2.8 GB Fits | 5.0 GB Fits | 3352 |
• real GGUF file size. Speed is a bandwidth ceiling (4800 GB/s), not a benchmark. Methodology.
More
GOT OCR2 0 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 (stepfun-ai/GOT-OCR2_0), NVIDIA specs. Data as of 2026-11-17.