What GPU do I need to run internlm/Intern-S1?

240.7B parameters, published in BF16. View on Hugging Face

240.7B
Parameters
BF16
Native precision
InternS1ForConditionalGeneration
Architecture
image-text-to-text
Pipeline

Intern-S1 is published by internlm on Hugging Face, with 19,518 downloads and 258 likes to date. It's a InternS1ForConditionalGeneration model built for image-text-to-text, published natively in BF16.

VRAM required & cheapest live GPU fit

Required VRAM = weight size at each precision times a flat 1.2 overhead for activations and fragmentation. The KV-cache grows with context and is not in that factor; it is listed per model below where the architecture is published. Full formula and assumptions: methodology.

PrecisionWeight sizeRequired VRAMCheapest live fitGPUs neededEst. $/hr (full fit)
BF16448.4 GB538.0 GBAMD MI300X3$7.89/hr
FP8 (quantized)224.2 GB269.0 GBRTX PRO 60003$4.13/hr
INT4 (quantized)112.1 GB134.5 GBAMD MI300X1$2.63/hr
cheaper alt.RTX A50006$1.06/hr

A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.

INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.

Cheapest way to run Intern-S1 at its published (BF16) precision: 3× AMD MI300X, at $2.63/hr per GPU ($7.89/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.

Intern-S1: KV cache by context length

The KV cache is the memory the attention layers hold for every token of context, on top of the weights. It is not part of the flat 1.2x overhead in the table above, grows with context length and with every concurrent request, and is why a long-context deployment needs more VRAM than the table shows.

Grouped-query attention: every layer caches keys and values for a few shared KV heads. Cached per token: 192,512 bytes at 16-bit.

ContextKV cache, one sequence
4K0.73 GB
32K5.9 GB
64K (model maximum)11.8 GB

Computed from the layer, head and window counts in the model's own configuration (internlm/Intern-S1), at 16 bits per cached value. Each concurrent sequence needs its own cache, on top of the weights.

Intern-S1: common questions

Can Intern-S1 run on a single GPU?

No. At BF16 it needs 538.0 GB of VRAM, more than a single 32 GB desktop card holds. The cheapest capable card in the live feed is a 192.0 GB AMD MI300X, and it takes 3 of them.

How many GPUs do I need to run Intern-S1?

3 at BF16. It needs 538.0 GB of VRAM and the cheapest capable live offer is a 192.0 GB AMD MI300X, so 3 of them come to $7.89/hr in total.

Does quantizing Intern-S1 lower the GPU bill?

Yes. At BF16 the cheapest live fit is 3 AMD MI300X cards at $7.89/hr. At INT4 (quantized) it drops to one AMD MI300X at $2.63/hr, provided a quantized checkpoint exists for it.

Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.

internlm models

Alternatives at this size

Other models for image-text-to-text within about a third of Intern-S1's 240.7B parameters, from other model lines.

More on Intern-S1

Related reading: AMD MI300X pricing and specs, How much VRAM you need for LLMs, Serving LLMs with vLLM, vLLM vs TensorRT-LLM vs SGLang, and Best GPU for LLM inference.

Your own GPU cloud accounts

One console for the GPU clouds you already rent from. Your keys stay on your machine.

$20 a month, flat

Connect your accounts

One console for the GPU clouds you already use.

Connect your accounts for $20 a month, flat. Your provider bills the GPU.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.