All compatibility checks

Compatibility check

Can you run Mistral Nemo Instruct on the H200 141GB?

Yes — runs in full precision

Yes. Mistral Nemo Instruct fits on the H200 141GB (141 GB) in full FP16/BF16 precision, using about 28.8 GB including a 0.6 GB KV cache at 4,096 tokens. You have comfortable headroom for longer prompts and modest batching.

Memory breakdown

Weights plus a 0.6 GB KV cache at 4,096tokens, against the card's 141 GB. Verdicts leave ~10% headroom for activations and fragmentation.

PrecisionWeightsKV cacheTotal% of 141 GBFit
FP16 / BF16full quality28.2 GB0.6 GB28.8 GB20%Fits
INT8 (8-bit)near-full quality14.1 GB0.6 GB14.7 GB10%Fits
INT4 (4-bit)GPTQ / AWQ / GGUF Q47 GB0.6 GB7.6 GB5%Fits

Planning estimates, not a substitute for profiling. Real usage varies with the inference runtime, batch size, and how much context you actually use — the KV cache grows linearly with prompt length.

GPUs that run Mistral Nemo Instruct

Cards where this model fits (at its best precision):

Go deeper

Frequently asked questions

Can the H200 141GB run Mistral Nemo Instruct?

Yes. In FP16 it uses about 28.8 GB, which fits the H200 141GB's 141 GB.

How much VRAM does Mistral Nemo Instruct need?

Approximately 28.2 GB in FP16, 14.1 GB in INT8, and 7 GB in 4-bit for the weights, plus a KV cache of about 0.6 GB at 4,096 tokens.

Does quantization let Mistral Nemo Instruct fit on the H200 141GB?

Yes. Dropping to FP16 / BF16 brings total usage to about 28.8 GB, which fits the 141 GB card with headroom for the KV cache.

What happens to memory with longer context?

The KV cache grows linearly with prompt length. At 4,096 tokens it is about 0.6 GB here; doubling the context roughly doubles that term, so long-context use can push a tight fit over the edge.