Can DeepSeek Coder V2 236B run on NVIDIA H200 141GB?

YES — With Q2_K

A79Great

○Estimated from fit model

DeepSeek Coder V2 236B needs ~165.6 GB VRAM. NVIDIA H200 141GB has 141.0 GB. With Q2_K quantization, expect ~66 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: HighStack: StandardBottleneck: Host offload

Operating mode

Choose the run profile you care about

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.

DeepSeek Coder V2 236B at Q4_K_M needs 217.6 GB — too much for NVIDIA H200 141GB (141.0 GB). Runs at Q2_K (165.6 GB) with low quality.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 217.6 GB, exceeds 141.0 GB available

217.6 GB required141.0 GB available

154% VRAM needed

76.6 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

31.5 tok/s

TTFT

6153 ms

Safe context

Memory

217.6 GB / 141.0 GB

Offload

40%

Memory breakdown

Weights144.0 GB

KV Cache58.6 GB

Runtime0.9 GB

Headroom14.1 GB

See how fast it feels

With memory offload — actual speed may be lower

See how fast it feelsDeepSeek Coder V2 236B on NVIDIA H200 141GB
1st promptCold start — includes initialization
>Write a Python function to sort a list
Here's a Python function that sorts a list using quicksort:

def quicksort(arr):
 if len(arr) <= 1:
 return arr
 pivot = arr[len(arr) // 2]
 left = [x for x in arr if x < pivot]
 middle = [x for x in arr if x == pivot]
 right = [x for x in arr if x > pivot]
 return quicksort(left) + middle + quicksort(right)

# Example usage
numbers = [3, 6, 8, 10, 1, 2, 1]
print(quicksort(numbers)) # [1, 1, 2, 3, 6, 8, 10]

The algorithm picks a pivot, partitions the list into smaller, equal, and larger elements, then recursively sorts each partition and concatenates the results.
2nd promptWarm — model ready, no init overhead
>Now add type hints and a docstring
def quicksort(arr: list[int]) -> list[int]:
 """Sort a list using the quicksort algorithm.

 Args:
 arr: List of integers to sort.

 Returns:
 A new sorted list.
 """
 if len(arr) <= 1:
 return arr
 pivot = arr[len(arr) // 2]
 left = [x for x in arr if x < pivot]
 middle = [x for x in arr if x == pivot]
 right = [x for x in arr if x > pivot]
 return quicksort(left) + middle + quicksort(right)

Added type hints for the input and return type, plus a Google-style docstring.
Estimated: 31.5 tok/s decode · 6.2s TTFT (warm) · 79 tok/s prefill

What limits this setup

It fits through host-memory offload, and offload is the main reason performance drops.

CPU or host-memory offload is active

About 10% of the working set spills out of accelerator memory, which usually hurts latency and sustained decode throughput.

Very little memory headroom

You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.

Best improvement path

Remove offload with more accelerator memory

Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Buy headroom, not only minimum fit

A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.

Increase host RAM if you keep offloading

This setup may need roughly 13.7 GB of extra host RAM just for the offloaded portion, before OS and other tools.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	F	Too heavy	39.9 tok/s	2644 ms	4K
Coding	F	Too heavy	31.5 tok/s	6153 ms	4K
Agentic Coding	F	Too heavy	21.2 tok/s	13265 ms	4K
Reasoning	F	Too heavy	31.5 tok/s	7271 ms	4K
RAG	F	Too heavy	21.2 tok/s	16581 ms	4K

Quantization options

How DeepSeek Coder V2 236B (236B params) fits at each quantization level on NVIDIA H200 141GB (141.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_KBest for your GPU	2	92.0 GB	Low	A84
Q3_K_S	3	115.6 GB	Low	F0
NVFP4	4	132.2 GB	Medium	F0
Q4_K_M	4	144.0 GB	Medium	F0
Q5_K_M	5	169.9 GB	High	F0
Q6_K	6	193.5 GB	High	F0
Q8_0	8	252.5 GB	Very High	F0
F16	16	483.8 GB	Maximum	F0

Get started

Copy-paste commands to run DeepSeek Coder V2 236B on your machine.

Run

lms load DeepSeek-Coder-V2-Instruct && lms server start

Upgrade options

Hardware that runs DeepSeek Coder V2 236B well

AMD Instinct MI350X 288GBBudget pick

288 GB VRAM (+147)8000 GB/s (+3200)

Makes the model fit on the accelerator instead of staying completely out of reach.109.3 tok/s decode

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

~$8,000 MSRP

👁 NVIDIA
B100 192GBBest value

192 GB VRAM (+51)8000 GB/s (+3200)

Makes the model fit on the accelerator instead of staying completely out of reach.84 tok/s decode

Makes the model fit on the accelerator instead of staying completely out of reach.

Raises estimated decode speed by about 167%.

~$35,000 MSRP

👁 NVIDIA
NVIDIA GB200 192GBNVIDIA upgrade

192 GB VRAM (+51)8000 GB/s (+3200)

Makes the model fit on the accelerator instead of staying completely out of reach.84 tok/s decode

Makes the model fit on the accelerator instead of staying completely out of reach.

Raises estimated decode speed by about 167%.

~$60,000 MSRP

Frequently asked questions

See all results for NVIDIA H200 141GB See all hardware for DeepSeek Coder V2 236B

URL: https://willitrunai.com/can-run/deepseek-coder-v2-236b-on-h200-141gb

⇱ DeepSeek Coder V2 236B on NVIDIA H200 141GB? No — Alternati…

Can DeepSeek Coder V2 236B run on NVIDIA H200 141GB?

YES — With Q2_K

Choose the run profile you care about

Memory breakdown

See how fast it feels

What limits this setup

Best improvement path

Performance by workload

Quantization options

Get started

Hardware that runs DeepSeek Coder V2 236B well

Frequently asked questions