LFM2.5-350M GGUF size and VRAM requirements
LiquidAI/LFM2.5-350M-GGUF is a compact language model with 0.35 billion parameters, built on the lfm2 architecture. It is released under the other license and has been downloaded 13,752 times.
To run LiquidAI/LFM2.5-350M-GGUF locally at a 4,096-token context, its quantized versions need between 1.25 GB (Q4_0, lowest quality) and 1.71 GB (BF16, highest quality) of memory, weights plus KV cache and a system margin included.
For most users the best balance is BF16, needing about 1.71 GB. That means LiquidAI/LFM2.5-350M-GGUF fits entirely in the VRAM of a 6 GB GPU or larger, running fully on the GPU.
GGUF file size and memory by quantization
Compare real GGUF weight sizes, estimated KV cache and total memory for Q4, Q5, Q8 and every quantization published in this repository.
| Quant. | Bits | Quality | Weights | KV | Total | Speed~ | Verdict |
|---|---|---|---|---|---|---|---|
| Q4_0 | 4.95 | Good | 0.2 GB | 0.25 GB | 1.25 GB | 1958.4 t/s | Fits in VRAM |
| Q4_K_M | 5.18 | Very good | 0.21 GB | 0.25 GB | 1.26 GB | 1873.0 t/s | Fits in VRAM |
| Q5_K_M | 5.88 | Very good | 0.24 GB | 0.25 GB | 1.29 GB | 1649.5 t/s | Fits in VRAM |
| Q6_K | 6.62 | Excellent | 0.27 GB | 0.25 GB | 1.32 GB | 1464.0 t/s | Fits in VRAM |
| Q8_0 | 8.56 | Excellent | 0.35 GB | 0.25 GB | 1.4 GB | 1132.6 t/s | Fits in VRAM |
| BF16 | 16.06 | Excellent | 0.66 GB | 0.25 GB | 1.71 GB | 603.7 t/s | Fits in VRAM |
| F16 | 16.06 | Excellent | 0.66 GB | 0.25 GB | 1.71 GB | 603.7 t/s | Fits in VRAM |
KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.
Frequently asked questions
How much VRAM do you need to run LiquidAI/LFM2.5-350M-GGUF?
You need about 1.71 GB of VRAM to run LiquidAI/LFM2.5-350M-GGUF entirely on the GPU using the BF16 quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.
Can I run LiquidAI/LFM2.5-350M-GGUF on an 8 GB GPU?
Yes. With 8 GB of VRAM you can run LiquidAI/LFM2.5-350M-GGUF fully on the GPU using BF16 (about 1.71 GB).
Can I run LiquidAI/LFM2.5-350M-GGUF on a 16 GB GPU?
Yes. With 16 GB of VRAM you can run LiquidAI/LFM2.5-350M-GGUF fully on the GPU using BF16 (about 1.71 GB).
Can I run LiquidAI/LFM2.5-350M-GGUF on a 24 GB GPU?
Yes. With 24 GB of VRAM you can run LiquidAI/LFM2.5-350M-GGUF fully on the GPU using BF16 (about 1.71 GB).
What is the best quantization for LiquidAI/LFM2.5-350M-GGUF?
If memory allows, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.