HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/nvidia/nvidia/Nemotron-Cascade-8B-Thinking

nvidia/Nemotron-Cascade-8B-Thinking

family nemotron · 2 variants · unknown

provider
nvidia
released
—
huggingface
https://huggingface.co/nvidia/Nemotron-Cascade-8B-Thinking
best scores
MMLU: 83.0MMLU-Pro: 75.1GPQA Diamond: 62.0Chatbot Arena: 85.8IFEval: 85.0IFBench: 34.4AIME 2024: 76.0AIME 2025: 67.3BFCL (function calling): 68.1

Best scores per benchmark

benchmarkcategorybest score
Chatbot Arenachat85.8
IFEvalinstruction-following85.0
MMLUknowledge83.0
AIME 2024math76.0
MMLU-Proknowledge75.1
BFCL (function calling)agentic68.1
AIME 2025math67.3
GPQA Diamondreasoning62.0
IFBenchinstruction-following34.4

Variants (2)

variantkindquantbitsvRAMdlorgyear
Nemotron-Cascade-8B-Thinkingoriginalbf1616.0440nvidia2026
nvidia_Nemotron-Cascade-8B-Thinking-GGUFggufgguf2.0428bartowski2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.