HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/LiquidAI/LiquidAI/LFM2.5-1.2B-Thinking

LiquidAI/LFM2.5-1.2B-Thinking

family misc · 4 variants · unknown

provider
LiquidAI
released
—
huggingface
https://huggingface.co/LiquidAI/LFM2.5-1.2B-Thinking
best scores
GPQA Diamond: 38.89MMLU-Pro: 44.35IFEval: 86.23IFBench: 47.33GSM8K: 64.52MATH-500: 63.2AIME 2025: 14.0

Best scores per benchmark

benchmarkcategorybest score
IFEvalinstruction-following86.23
GSM8Kmath64.52
MATH-500math63.2
IFBenchinstruction-following47.33
MMLU-Proknowledge44.35
GPQA Diamondreasoning38.89
AIME 2025math14.0

Variants (4)

variantkindquantbitsvRAMdlorgyear
LFM2.5-1.2B-Thinking-GGUFggufgguf16.01,882unsloth2026
LFM2.5-1.2B-Thinking-GGUFggufgguf16.01,865lmstudio-community2026
LFM2.5-1.2B-Thinking-MLX-8bitmlxmlx1,644lmstudio-community2026
LFM2.5-1.2B-Thinking-MLX-bf16mlxmlx287lmstudio-community2026
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.