HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/tiiuae/tiiuae/Falcon-H1-1.5B-Instruct

tiiuae/Falcon-H1-1.5B-Instruct

family falcon · 4 variants · unknown

provider
tiiuae
released
—
huggingface
https://huggingface.co/tiiuae/Falcon-H1-1.5B-Instruct
best scores
BBH: 46.47TruthfulQA: 45.98HellaSwag: 63.33MMLU: 64.13GSM8K: 74.98MATH-500: 74.0GPQA: 26.34GPQA Diamond: 35.19MMLU-Pro: 37.8HumanEval: 68.29MBPP: 64.81LiveCodeBench: 17.61IFEval: 80.66AlpacaEval: 28.18MT-Bench: 8.46

Best scores per benchmark

benchmarkcategorybest score
IFEvalinstruction-following80.66
GSM8Kmath74.98
MATH-500math74.0
HumanEvalcoding68.29
MBPPcoding64.81
MMLUknowledge64.13
HellaSwagknowledge63.33
BBHreasoning46.47
TruthfulQAsafety45.98
MMLU-Proknowledge37.8
GPQA Diamondreasoning35.19
AlpacaEvalchat28.18
GPQAreasoning26.34
LiveCodeBenchcoding17.61
MT-Benchchat8.46

Variants (4)

variantkindquantbitsvRAMdlorgyear
Falcon-H1-1.5B-Instruct-GGUFggufgguf16.0703tiiuae2025
Falcon-H1-1.5B-Instruct-GGUFggufgguf16.0150unsloth2025
Falcon-H1-1.5B-Instruct-GPTQ-Int4gptqgptq97tiiuae2025
Falcon-H1-1.5B-Instruct-GPTQ-Int8gptqgptq60tiiuae2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.