Raon-1-mini-preview GPU Requirements: VRAM & Cheapest GPU
Raon-1-mini-preview has about 28.8B parameters. See exactly how much GPU memory it needs at FP16, INT8, and INT4, and the cheapest GPU to run it, with live hourly pricing from 5+ data center partners.
- Parameters
- 28.8B
- Min VRAM (INT4)
- 16 GB
- Cheapest (INT4)
- $0.58/hr
- Deploy
- < 2 min
To run Raon-1-mini-preview for inference at FP16, you need roughly 63 GB of VRAM. The cheapest fit on Spheron is 2x RTX 5090 32GB at about $1.36/hr. Quantize to INT4 to run it on a smaller, cheaper GPU.
VRAM required
Estimated peak VRAM including weights, activations, and KV cache. Add 10% headroom for production traffic.
- 01$0.68/hr$1.36/hr2× RTX 5090 32GBCHEAPESTBlackwell · GDDR7
- 02$1.43/hr$1.43/hr1× A100 80GBAmpere · HBM2e
- 03$0.78/hr$1.56/hr2× RTX 6000 Ada 48GBAda Lovelace · GDDR6 ECC
- 04$0.96/hr$1.92/hr2× L40S 48GBAda Lovelace · GDDR6
- 05$0.58/hr$2.32/hr4× RTX 4090 24GBAda Lovelace · GDDR6X
Live pricing aggregated from 5+ data center partners. Per-minute billing, no commitments.
VRAM required to run Raon-1-mini-preview
Estimated peak VRAM at context length 4,096 and batch size 1, including weights, activations, and KV cache. Quantizing to INT8 (Q8) or INT4 (Q4) cuts memory roughly in half and in quarter.
| Precision | Inference | LoRA fine-tune | Full fine-tune |
|---|---|---|---|
| FP16 | 63 GB | 94 GB | 251 GB |
| INT8 | 31 GB | 47 GB | 125 GB |
| INT4 | 16 GB | 24 GB | 63 GB |
Cheapest GPU to run Raon-1-mini-preview by precision
63GB
Full precision. Best quality, highest memory.
31GB
8-bit quantized. ~2x smaller, minimal quality loss.
16GB
4-bit quantized. ~4x smaller, runs on smaller GPUs.
Inference vs fine-tuning Raon-1-mini-preview
Inference only holds the model weights plus a KV cache, so it is the cheapest setup. LoRA fine-tuning adds a small adapter and roughly 50% more memory. Full fine-tuning holds gradients and optimizer state on top of the weights, which is about 4x the inference footprint, so it often needs multiple GPUs even when inference fits on one. For Raon-1-mini-preview, an on-demand RTX 5090 32GB instance covers inference and LoRA, while a full fine-tune needs several times that memory and often spans multiple GPUs. Check the live GPU pricing for current rates.
Similar models
Compare GPU requirements for models in the same class.
28.9Bibm-granite/granite-4.1-30b
29.3Bibm-granite/granite-4.2-30b
27.8BOBLITERATUS/Qwen3.8-27B-OBLITERATED
27.8BAEON-7/Qwen3.8-27B-AEON-ULTIMATE-UNCENSORED-BF16
30.1Btencent/Hy-Memory-30B-A3B
27.4Bprism-ml/Ternary-Bonsai-2-27B-mlx-2bit
Raon-1-mini-preview GPU questions
VRAM, GPU choice and cost questions for running Raon-1-mini-preview on cloud GPUs.
Raon-1-mini-preview has about 28.8B parameters. At FP16 it needs roughly 63 GB of VRAM for inference, including weights, activations, and KV cache. Quantized to INT4 that drops to around 16 GB. Leave about 10% headroom for production traffic.
For FP16 inference, the cheapest fit on Spheron is 2x RTX 5090 32GB at about $1.36/hr. If you quantize to INT4, RTX 4090 24GB at about $0.58/hr runs it for less. Pricing is aggregated live from 5+ data center partners with per-minute billing.
LoRA fine-tuning adds roughly 50% on top of inference memory, so it usually fits the same class of GPU. Full fine-tuning holds gradients and optimizer state and needs about 4x the inference VRAM, which often means multiple GPUs for Raon-1-mini-preview. The VRAM matrix above shows the exact estimate for each setup.
We read the parameter count directly from the model's safetensors metadata on HuggingFace, then estimate peak VRAM from weights, activations, KV cache, and framework overhead at your chosen precision. The estimate lands within about 15% of real-world use for most transformer models.
Quantized to INT4, Raon-1-mini-preview needs about 16 GB of VRAM, so it fits on a 24 GB RTX 4090 with room for a small batch. At FP16 it needs roughly 63 GB, which usually means a data center GPU. The precision picks above list the cheapest GPU that fits each setup.
Yes. At FP16 Raon-1-mini-preview needs about 63 GB of VRAM, which fits one 80 GB H100 with about 17 GB left for longer contexts and larger batches. The precision picks above show the cheapest GPU that fits each setup on Spheron.
Raon-1-mini-preview needs one 80 GB GPU at FP16 (about 63 GB of VRAM), one 80 GB GPU at INT8 (31 GB) and one 80 GB GPU at INT4 (16 GB). All of these fit inside a single 8-GPU node, so no multi-node setup is needed for inference. Smaller cards such as 48 GB or 24 GB GPUs need proportionally more of them.
Raon-1-mini-preview needs about 31 GB of VRAM at INT8, roughly half its FP16 footprint of 63 GB. The cheapest INT8 fit on Spheron is RTX 5090 32GB at about $0.68/hr. INT8 is the usual middle ground: half the memory of FP16, with a smaller quality drop than 4-bit quantization.
Running Raon-1-mini-preview around the clock at FP16 costs about $993 a month on 2x RTX 5090 32GB (730 hours at $1.36/hr). Quantized to INT4 it costs about $423 a month on RTX 4090 24GB (730 hours at $0.58/hr). Spheron bills per minute, so stopping the instance when it is idle cuts that figure directly.
Raon-1-mini-preview needs about 63 GB at FP16, 31 GB at INT8 and 16 GB at INT4. Use FP16 when output quality matters most or you plan to fine-tune, INT8 for most production inference, and INT4 when cost or GPU size is the constraint: the cheapest INT4 fit is RTX 4090 24GB at about $0.58/hr.