# NVIDIA H200

8 Hacker News commenters recommended NVIDIA H200 and 5 criticized it, November 2025 to November 2026. It was recommended in 5 of 13 months.

**Verdict:** Datacenter GPU that people recommend for running large open-source models unquantized, mostly by renting cloud instances by the hour. Critics call it overpriced, with limited concurrency and throughput.

## Why people recommend it

- Runs large open models unquantized (2 comments): "to run unquantized open source models with abliteration/heretic transformations" ([source](https://news.ycombinator.com/item?id=49969722)) "I ran glm 5.2 on rented 8x h200 it could only do 2x concurrency at a cost of $40 an hour" ([source](https://news.ycombinator.com/item?id=48619736))
- Renting by the hour is practical (2 comments): "It's a depreciating asset. You rent one." ([source](https://news.ycombinator.com/item?id=48561450)) "spin up 4xH200s at ~$9/hr on spot" ([source](https://news.ycombinator.com/item?id=49667917))
- Faster and cheaper than H100 setups (2 comments): "8xH200s will be a lot cheaper than 16xH100s, and faster too." ([source](https://news.ycombinator.com/item?id=46778152)) "H200 is a major step up over H100." ([source](https://news.ycombinator.com/item?id=46970758))

## What people warn about

- Overpriced for what it delivers (2 comments): "H200's are not that cost effective and saturate too fast while having poorer efficiency" ([source](https://news.ycombinator.com/item?id=48684770)) "The datacenter GPUs sell at an 80% margin. They're incredibly overpriced." ([source](https://news.ycombinator.com/item?id=48259539))
- Limited concurrency and throughput (3 comments): "I ran glm 5.2 on rented 8x h200 it could only do 2x concurrency at a cost of $40 an hour" ([source](https://news.ycombinator.com/item?id=48619736)) "performance/throughput will be roughly halved (closer to 2/3) and free context will be more limited with 8xH200" ([source](https://news.ycombinator.com/item?id=46780738)) "H200's are not that cost effective and saturate too fast while having poorer efficiency" ([source](https://news.ycombinator.com/item?id=48684770))
- Needs a rack and is expensive (2 comments): "That GPU costs 25k which means you really should have a rack to put it in. It's not realistic." ([source](https://news.ycombinator.com/item?id=48560666)) "The datacenter GPUs sell at an 80% margin. They're incredibly overpriced." ([source](https://news.ycombinator.com/item?id=48259539))

## Alternatives hackers recommend

- [NVIDIA H100](https://hackerapproved.com/p/nvidia-h100.md): 20 recommend
- [NVIDIA Tesla V100](https://hackerapproved.com/p/nvidia-tesla-v100.md): 16 recommend
- [AMD Instinct MI355X](https://hackerapproved.com/p/amd-instinct.md): 14 recommend
- [Nvidia A100](https://hackerapproved.com/p/nvidia-a100-gpu.md): 10 recommend
- [Intel D270 GPU AI Accelerator](https://hackerapproved.com/p/intel-d270-gpu-ai-accelerator.md): 10 recommend

## Where to get it

- [Visit the official site](https://www.nvidia.com/en-us/data-center/h200/)

## Cite as

Hacker Approved, "NVIDIA H200", https://hackerapproved.com/p/nvidia-h200/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).