LFM2.5-24B-A2B GPU Requirements: VRAM & Cheapest GPU
LFM2.5-24B-A2B has about 24.0B 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.
To run LFM2.5-24B-A2B for inference at FP16, you need roughly 52 GB of VRAM. The cheapest fit on Spheron is A100 80GB from $1.26/hr. Quantize to INT4 to run it on a smaller, cheaper GPU.
Estimated peak VRAM including weights, activations, and KV cache. Add 10% headroom for production traffic.
- 01$1.43/hr$1.43/hr1× A100 80GBCHEAPESTAmpere · HBM2e
- 02$0.78/hr$1.56/hr2× RTX 6000 Ada 48GBAda Lovelace · GDDR6 ECC
- 03$0.78/hr$1.56/hr2× RTX 5090 32GBBlackwell · GDDR7
- 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 LFM2.5-24B-A2B
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 | 52 GB | 78 GB | 209 GB |
| INT8 | 26 GB | 39 GB | 105 GB |
| INT4 | 13 GB | 20 GB | 52 GB |
Cheapest GPU to run LFM2.5-24B-A2B by precision
Full precision. Best quality, highest memory.
8-bit quantized. ~2x smaller, minimal quality loss.
4-bit quantized. ~4x smaller, runs on smaller GPUs.
Inference vs fine-tuning LFM2.5-24B-A2B
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 LFM2.5-24B-A2B, an on-demand A100 80GB 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.
Deployment guideDeploy LFM2 step by stepHands-on production setup, GPU configs, and benchmarks for LFM2.5-24B-A2B.Read guideSimilar models
Compare GPU requirements for models in the same class.
LFM2.5-24B-A2B GPU questions
LFM2.5-24B-A2B has about 24.0B parameters. At FP16 it needs roughly 52 GB of VRAM for inference, including weights, activations, and KV cache. Quantized to INT4 that drops to around 13 GB. Leave about 10% headroom for production traffic.
For FP16 inference, the cheapest fit on Spheron is A100 80GB from $1.26/hr. If you quantize to INT4, RTX 4090 24GB from $0.72/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 LFM2.5-24B-A2B. 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, LFM2.5-24B-A2B needs about 13 GB of VRAM, so it fits on a 24 GB RTX 4090 with room for a small batch. At FP16 it needs roughly 52 GB, which usually means a data center GPU. The precision picks above list the cheapest GPU that fits each setup.