NVIDIA H20 96GBBudget pick96 GB VRAM (+16)4000 GB/s (+1000)
C
Raises estimated decode speed by about 33%.73.8 tok/s decode
Raises estimated decode speed by about 33%.
~$12,000 MSRP
BaichuanMed OCR 72B i1 needs ~61.6 GB VRAM. NVIDIA H800 80GB has 80.0 GB. With Q4_K_M quantization, expect ~55 tok/s.
Operating mode
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
Fit status
Runs well
Decode
55.3 tok/s
TTFT
3499 ms
Safe context
51K
Memory
61.6 GB / 80.0 GB
This setup is broadly balanced for this model.
No major red flags
This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | C | Runs well | 55.3 tok/s | 1909 ms | 51K |
| Coding | C | Runs well | 55.3 tok/s | 3499 ms | 51K |
| Agentic Coding | C | Tight fit | 55.3 tok/s | 5090 ms | 51K |
| Reasoning | C | Runs well | 55.3 tok/s | 4135 ms | 51K |
| RAG | C | Tight fit | 55.3 tok/s | 6362 ms | 51K |
Inference speed
Estimated decode speed (tokens/sec) for BaichuanMed OCR 72B i1 at Q4_K_M across popular GPUs and Apple Silicon, including multi-GPU rigs, using the fastest local runtime per device. Fastest is MacBook Pro M4 Max 128GB at ~14 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.
| GPU / Mac | Memory | Quant | Speed (tok/s) | Fits? |
|---|---|---|---|---|
MacBook Pro M4 Max 128GB | 128 GB | Q4_K_M | 13.7 | Fits |
2× RX 7900 XTX 24GB | 48 GB | Q4_K_M | 13.4 | Too big |
Mac Studio M3 Ultra 256GB | 256 GB | Q4_K_M | 12.7 | Fits |
Mac Studio M2 Ultra 128GB | 128 GB | Q4_K_M | 10.6 | Fits |
Mac Studio M1 Ultra 128GB | 128 GB | Q4_K_M | 10.0 | Fits |
MacBook Pro M4 Max 64GB | 64 GB | Q4_K_M | 9.4 | Too big |
| 2× RTX 4090 24GB | 48 GB | Q4_K_M | 7.1 | Too big |
| RTX 5090 32GB | 32 GB | Q4_K_M | 6.5 | Too big |
| 2× RTX 3090 24GB | 48 GB | Q4_K_M | 6.5 | Too big |
| 4× RTX 3060 12GB | 48 GB | Q4_K_M | 5.7 | Too big |
MacBook Pro M4 Pro 48GB | 48 GB | Q4_K_M | 4.4 | Too big |
MacBook Pro M3 Max 64GB | 64 GB | Q4_K_M | 3.8 | Too big |
MacBook Pro M1 Max 64GB | 64 GB | Q4_K_M | 3.4 | Too big |
| RTX 4090 24GB | 24 GB | Q4_K_M | 2.6 | Too big |
RX 7900 XTX 24GB | 24 GB | Q4_K_M | 2.4 | Too big |
| RTX 3090 24GB | 24 GB | Q4_K_M | 2.2 | Too big |
| RTX 4080 Super 16GB | 16 GB | Q4_K_M | 2.1 | Too big |
| RTX 4070 12GB | 12 GB | Q4_K_M | 2.0 | Too big |
| RTX 3060 12GB | 12 GB | Q4_K_M | 2.0 | Too big |
| RTX 4060 8GB | 8 GB | Q4_K_M | 2.0 | Too big |
Estimates for single-stream decoding at Q4_K_M; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.
How BaichuanMed OCR 72B i1 (72B params) fits at each quantization level on NVIDIA H800 80GB (80.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 28.1 GB | Low | C43 |
Q3_K_S | 3 | 35.3 GB | Low | C45 |
NVFP4 | 4 | 40.3 GB | Medium | C47 |
Q4_K_M | 4 | 43.9 GB | Medium | C47 |
Q5_K_M | 5 | 51.8 GB | High | C47 |
Q6_KBest for your GPU | 6 | 59.0 GB | High | C47 |
Q8_0 | 8 | 77.0 GB | Very High | F0 |
F16 | 16 | 147.6 GB | Maximum | F0 |
Copy-paste commands to run BaichuanMed OCR 72B i1 on your machine.
Run
lms load hf-mradermacher--baichuanmed-ocr-72b-i1-gguf && lms server startUpgrade options
NVIDIA H20 96GBBudget pickRaises estimated decode speed by about 33%.
~$12,000 MSRP
NVIDIA GH200 96GBBest valueRaises estimated decode speed by about 33%.
~$30,000 MSRP
Yes, NVIDIA H800 80GB can run BaichuanMed OCR 72B i1 with a C grade (Runs well). Expected decode speed: 55.3 tok/s.
BaichuanMed OCR 72B i1 (72B parameters) requires approximately 61.6 GB of memory with Q4_K_M quantization.
The recommended quantization for BaichuanMed OCR 72B i1 is Q4_K_M, which balances quality and memory efficiency.
On NVIDIA H800 80GB, BaichuanMed OCR 72B i1 achieves approximately 55.3 tokens per second decode speed with a time-to-first-token of 3499ms using Q4_K_M quantization.
For coding workloads, BaichuanMed OCR 72B i1 on NVIDIA H800 80GB receives a C grade with 55.3 tok/s and 51K context.
On NVIDIA H800 80GB, BaichuanMed OCR 72B i1 can safely use up to 51K tokens of context. The model's official context limit is —, but available memory constrains the safe maximum.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/hf-mradermacher--baichuanmed-ocr-72b-i1-gguf-on-h800-80gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview: