HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/LiquidAI/LiquidAI/LFM2-8B-A1B

LiquidAI/LFM2-8B-A1B

family misc · 3 variants · unknown

provider
LiquidAI
released
—
huggingface
https://huggingface.co/LiquidAI/LFM2-8B-A1B
best scores
MMLU: 64.84MMLU-Pro: 37.42GPQA: 29.29IFEval: 77.58IFBench: 25.85GSM8K: 84.38MATH-500: 74.2MATH: 62.38MGSM: 72.4HumanEval: 69.51

Best scores per benchmark

benchmarkcategorybest score
GSM8Kmath84.38
IFEvalinstruction-following77.58
MATH-500math74.2
MGSMmath72.4
HumanEvalcoding69.51
MMLUknowledge64.84
MATHmath62.38
MMLU-Proknowledge37.42
GPQAreasoning29.29
IFBenchinstruction-following25.85

Variants (3)

variantkindquantbitsvRAMdlorgyear
LFM2-8B-A1B-GGUFggufgguf16.02,332unsloth2025
LiquidAI_LFM2-8B-A1B-GGUFggufgguf2.01,038bartowski2025
LFM2-8B-A1Boriginalbf1616.0122unsloth2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.