HF Model & Benchmark Matrix

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

tiiuae/Falcon-H1-1.5B-Deep-Instruct

family falcon · 4 variants · unknown

provider
tiiuae
released
—
huggingface
https://huggingface.co/tiiuae/Falcon-H1-1.5B-Deep-Instruct
best scores
BBH: 54.43TruthfulQA: 50.48HellaSwag: 65.54MMLU: 67.3GSM8K: 82.34MATH-500: 77.8GPQA: 33.22GPQA Diamond: 40.57MMLU-Pro: 41.89HumanEval: 73.78MBPP: 68.25LiveCodeBench: 23.87IFEval: 83.5AlpacaEval: 27.12MT-Bench: 8.53

Best scores per benchmark

benchmarkcategorybest score
IFEvalinstruction-following83.5
GSM8Kmath82.34
MATH-500math77.8
HumanEvalcoding73.78
MBPPcoding68.25
MMLUknowledge67.3
HellaSwagknowledge65.54
BBHreasoning54.43
TruthfulQAsafety50.48
MMLU-Proknowledge41.89
GPQA Diamondreasoning40.57
GPQAreasoning33.22
AlpacaEvalchat27.12
LiveCodeBenchcoding23.87
MT-Benchchat8.53

Variants (4)

variantkindquantbitsvRAMdlorgyear
Falcon-H1-1.5B-Deep-Instruct-GGUFggufgguf16.0851tiiuae2025
Falcon-H1-1.5B-Deep-Instruct-GGUFggufgguf16.0379unsloth2025
Falcon-H1-1.5B-Deep-Instructoriginalbf1616.065unsloth2025
Falcon-H1-1.5B-Deep-Instruct-GPTQ-Int4gptqgptq54tiiuae2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.