microsoft/Phi-3-medium-4k-instruct
14B params · est. 8.4 GB Q4
NVIDIA B200 192GB: likely fitNVIDIA H200 141GB: likely fit
Head-to-head
NVIDIA B200 192GB vs NVIDIA H200 141GB
Dedicated comparison page for two real hardware profiles. This is built from your benchmark database, trust metadata, and buyer flow instead of generic spec-sheet comparisons.
Device profile
NVIDIA B200 192GB
Compared here against NVIDIA H200 141GB.
Curated Aggregate·2026-02-22192GB / $40k
Best LLM
520 tok/s
VRAM
192 GB
TDP
1000 W
Rating
9.5
Device profile
NVIDIA H200 141GB
405B-class models at FP8 with context headroom — the capacity play for frontier open-weight models
Curated Aggregate·2026-04-20141GB / $30k
Best LLM
540 tok/s
VRAM
141 GB
TDP
700 W
Rating
9.4
Quick verdict
NVIDIA B200 192GB scores 9.5 while NVIDIA H200 141GB scores 9.4 on MyAIHardware's composite rating.
NVIDIA B200 192GB wins VRAM capacity.
NVIDIA H200 141GB is the lower-power path.
On shared workload evidence, Llama 3 8B Q4 is benchmarked at 360 tok/s for NVIDIA B200 192GB and 215 tok/s for NVIDIA H200 141GB.
| Metric | NVIDIA B200 192GB | NVIDIA H200 141GB |
|---|---|---|
| MyAI rating | 9.5 | 9.4 |
| Best LLM | 520 tok/s | 540 tok/s |
| VRAM | 192 GB | 141 GB |
| TDP | 1000 W | 700 W |
| MSRP | $40k | $30k |
| Workloads | 12 | 11 |
Shared benchmark rows
Llama 3 8B Q4
Llama 3 8B Q4
NVIDIA B200 192GB
360 tok/s
Q4_K_M / 192GB / 2025-03-20
NVIDIA H200 141GB
215 tok/s
Q4_K_M / 141GB / 2025-02-04
Mistral 7B Q4
Mistral 7B
NVIDIA B200 192GB
415 tok/s
Q4_K_M / 192GB / 2025-01-08
NVIDIA H200 141GB
305 tok/s
Q4_K_M / 141GB / 2026-01-28
Gemma 2 9B Q4
Gemma 2 9B
NVIDIA B200 192GB
332 tok/s
Q4_K_M / 192GB / 2025-02-12
NVIDIA H200 141GB
220 tok/s
Q4_K_M / 141GB / 2024-07-30
DeepSeek-R1 Distill 7B
DeepSeek-R1 7B
NVIDIA B200 192GB
480 tok/s
Q4_K_M / 192GB / 2026-02-04
NVIDIA H200 141GB
305 tok/s
Q4_K_M / 141GB / 2026-04-20
Qwen 2.5 14B Q4
Qwen 2.5 14B
NVIDIA B200 192GB
285 tok/s
Q4_K_M / 192GB / 2026-01-15
NVIDIA H200 141GB
178 tok/s
Q4_K_M / 141GB / 2024-12-15
Llama 3 70B Q4
Llama 3 70B Q4
NVIDIA B200 192GB
138 tok/s
Q4_K_M / 192GB / 2024-12-26
NVIDIA H200 141GB
84.0 tok/s
Q4_K_M / 141GB / 2024-04-06
External quality layer
Best open models likely to fit this class
This shortlist cross-references the external OpenEvals snapshot with an approximate Q4-class VRAM estimate. Use it to sanity-check whether the hardware you are comparing can host strong open models, not just benchmark toy workloads.
91.0
Qwen/Qwen2-72B
73B params · est. 43.6 GB Q4
NVIDIA B200 192GB: likely fitNVIDIA H200 141GB: likely fit
microsoft/Phi-3.5-mini-instruct
3.8B params · est. 2.3 GB Q4
NVIDIA B200 192GB: likely fitNVIDIA H200 141GB: likely fit
internlm/internlm2_5-7b-chat
7.7B params · est. 4.6 GB Q4
NVIDIA B200 192GB: likely fitNVIDIA H200 141GB: likely fit