What GPU do I need to run meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8?
A 402B (MoE) vision-language model that reads images alongside text. 401.6B parameters, published in F8_E4M3. View on Hugging FaceGated
Llama-4-Maverick-17B-128E-Instruct-FP8 is published by meta-llama on Hugging Face, with 52,485 downloads and 181 likes to date. It's a Llama4ForConditionalGeneration model built for image-text-to-text, published natively in F8_E4M3, and gated: you'll need to accept the model's terms on Hugging Face before downloading weights.
What Llama-4-Maverick-17B-128E-Instruct-FP8 is
Llama-4-Maverick-17B-128E-Instruct-FP8 is a 402B-parameter mixture-of-experts vision-language model published by Meta on Hugging Face. It is released under Custom license.
License note: a lab-specific license (tagged "other" on Hugging Face); read the model card's own license section before commercial use. Facts in this section are sourced from Llama-4-Maverick-17B-128E-Instruct-FP8's Hugging Face model card, not benchmarked by Aquanode.
What it's used for
- Image understanding and captioning
- Visual question answering
- Document/OCR-style reading
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.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| FP8 (native) | 374.1 GB | 448.9 GB | RTX PRO 6000 | 5 | $6.05/hr |
| INT4 (quantized) | 187.0 GB | 224.4 GB | RTX A6000 | 5 | $1.81/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 Llama-4-Maverick-17B-128E-Instruct-FP8 at its published (F8_E4M3) precision: 5× RTX PRO 6000, at $1.21/hr per GPU ($6.05/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
Llama-4-Maverick-17B-128E-Instruct-FP8: 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.
KV-cache size for Llama-4-Maverick-17B-128E-Instruct-FP8 is not published for this architecture, so no figure is shown.
Llama-4-Maverick-17B-128E-Instruct-FP8: common questions
Can Llama-4-Maverick-17B-128E-Instruct-FP8 run on a single GPU?
No. At FP8 (native) it needs 448.9 GB of VRAM, more than a single 32 GB desktop card holds. The cheapest capable card in the live feed is a 95.0 GB RTX PRO 6000, and it takes 5 of them.
Do I need approval to download Llama-4-Maverick-17B-128E-Instruct-FP8?
Yes. meta-llama gates this repository on Hugging Face, so you have to accept its terms with a signed-in account before the weights will download. It does not change the 448.9 GB the model needs once you have them.
Is Llama-4-Maverick-17B-128E-Instruct-FP8 already quantized?
Yes. It is published in FP8, one byte per parameter, so the 448.9 GB figure is already a quantized footprint rather than a full-precision one. Only INT4 goes below it, at 224.4 GB. A GPU without FP8 tensor cores cannot run it as published, which is why cards here are matched on precision support and not on VRAM alone.
How many GPUs do I need to run Llama-4-Maverick-17B-128E-Instruct-FP8?
5 at FP8 (native). It needs 448.9 GB of VRAM and the cheapest capable live offer is a 95.0 GB RTX PRO 6000, so 5 of them come to $6.05/hr in total.
How to run Llama-4-Maverick-17B-128E-Instruct-FP8
Run Llama-4-Maverick-17B-128E-Instruct-FP8 with vLLM
Generic example, not from the model's own docs: adjust flags (quantization, context length, parallelism) for your setup.
vllm serve meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8 --tensor-parallel-size 5Deploy Llama-4-Maverick-17B-128E-Instruct-FP8 on Aquanode
Aquanode has no one-click deploy template for Llama-4-Maverick-17B-128E-Instruct-FP8; you install the inference engine yourself with the commands below. Aquanode sells GPU pods billed per second, not a hosted inference API.
- Launch a bare GPU pod sized to the requirement above (5× RTX PRO 6000 or larger).
- Open a terminal on the pod, or save one of the commands above as a startup script so it runs automatically the first time the pod boots.
- Run the command and connect to the resulting endpoint.
Launch a GPU podWeight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More Llama 4 models
- Llama-4-Scout-17B-16E-Instruct (108.6B, BF16)
Alternatives at this size
Other models for image-text-to-text within about a third of Llama-4-Maverick-17B-128E-Instruct-FP8's 401.6B parameters, from other model lines.
- GLM-5.3-Flash (321.3B, F8_E4M3)
- MiniMax-M3 (427.0B, BF16)
- Qwen3.5-397B-A17B (403.4B, BF16)
- step3 (321.0B, BF16)
More on Llama-4-Maverick-17B-128E-Instruct-FP8
Fits on a 288 GB GPU at INT4: every model that fits in 288 GB.
Related reading: RTX PRO 6000 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.
