Compare LLM Models

Select 2-5 models to compare side by side.

Select models to compare 0 / 5 selected

Top downloads

🤖 DeepSeek-V4-Pro-0813 deepseek-ai 🤖 GLM-5.3 zai-org 🤖 Kimi-K3 moonshotai

Select 2-5 models

Attribute	🤖 LFM2.5-1.2B-Instruct-ONNX LiquidAI
Company	LiquidAI
Release Date	2026-01-05
Parameters (B)	1.2
Architecture	LFM2.5 Hybrid
Context Window	32768
Input Modalities	text
Open Source	Yes
License	other
Score	5
VRAM (GB)	1.2
Compute	ONNX Runtime on CPU, GPU, or NPU; ~1.2 GB VRAM (Q4, recommended), ~2.4 GB (FP16), or ~1.7 GB (Q8, server only); WebGPU supports Q4 and FP16
Benchmark: AIME-2025	14
Benchmark: GPQA	38.9
Benchmark: GSM8K	64.5
Benchmark: IFEval	86.2
Benchmark: MATH-500	63.2
Benchmark: MMLU-Pro	44.4
Pricing (Input)	—
Pricing (Output)	—