Dataset · google
IFEval instruction-following benchmark
google/IFEval
Around 500 verifiable instructions for evaluating how well LLMs follow explicit constraints.
What's inside
A prompt dataset from the Instruction-Following Eval (IFEval) benchmark for large language models. Contains around 500 verifiable instructions, such as writing over 400 words or mentioning AI at least three times. Execution verification relies on heuristics. English language.
What you can do with it
- Evaluate LLMs on instruction following
- Compare base and fine-tuned models
- Test LoRA after SFT
- Tasks
- Text generation
- Languages
- English
- Size
- up to 1K rows
- Downloads this month
- 190.6K
- Likes
- 190
Models trained on it
- mradermacher/Nova-Qwen2.5-14B-Secure-Coder-GGUF 969
- Lautloserspieler/Nova-Qwen2.5-14B-Secure-CoderText generation 665
- rico03/Qwopus-GLM-9B-DualReason-Distilled-GGUFText generation 280
- veritiana-ai/prompt-task-complexity-classifierText classification 9
- moro72842/Sovereign-Hardened-V1 0
- Dhruv1000/qwen35-logra-ifbenchReinforcement learning 0
How to load it
from datasets import load_dataset
ds = load_dataset("google/IFEval")