MM IF-Eval

A challenging multimodal instruction-following benchmark that includes both compose-level constraints for output responses and perception-level constraints tied to input images, with comprehensive evaluation pipeline.

Pixtral-12B from Mistral AI currently leads the MM IF-Eval leaderboard with a score of 0.527 across 1 evaluated AI models.

Paper
About this benchmark

What MM IF-Eval measures

MM IF-Eval is a multimodal benchmark that evaluates large language models on multimodal, reasoning, and structured output tasks. LLM Stats tracks 1 model on this benchmark, with a maximum possible score of 1. Current average across reported models is 0.5, with the leader reaching 0.5.

Compare leaders on the best AI for multimodal, best AI for reasoning and best AI for structured output leaderboards.

Publication

Paper
MM-IFEngine: Towards Multimodal Instruction Following
Authors
Shengyuan Ding, Shenxi Wu, Xiangyu Zhao, Yuhang Zang, and 6 others
Published

Abstract

The Instruction Following (IF) ability measures how well Multi-modal Large Language Models (MLLMs) understand exactly what users are telling them and whether they are doing it right. Existing multimodal instruction following training data is scarce, the benchmarks are simple with atomic instructions, and the evaluation strategies are imprecise for tasks demanding exact output constraints. To address this, we present MM-IFEngine, an effective pipeline to generate high-quality image-instruction pairs. Our MM-IFEngine pipeline yields large-scale, diverse, and high-quality training data MM-IFInstruct-23k, which is suitable for Supervised Fine-Tuning (SFT) and extended as MM-IFDPO-23k for Direct Preference Optimization (DPO). We further introduce MM-IFEval, a challenging and diverse multi-modal instruction-following benchmark that includes (1) both compose-level constraints for output responses and perception-level constraints tied to the input images, and (2) a comprehensive evaluation pipeline incorporating both rule-based assessment and judge model. We conduct SFT and DPO experiments and demonstrate that fine-tuning MLLMs on MM-IFInstruct-23k and MM-IFDPO-23k achieves notable gains on various IF benchmarks, such as MM-IFEval (+10.2$\%$), MIA (+7.6$\%$), and IFEval (+12.3$\%$). We have fully open-sourced the datasets (both SFT and DPO), evaluation code and training scripts at https://github.com/SYuan03/MM-IFEngine.

Mistral AIPixtral-12B leads with 52.7%.

Progress Over Time

Interactive timeline showing model performance evolution on MM IF-Eval

State-of-the-art frontier
Open
Proprietary

MM IF-Eval Leaderboard

1 models
ContextCostLicense
1
Mistral AI
Mistral AI
12B
Notice missing or incorrect data?

FAQ

Common questions about MM IF-Eval.

What is the MM IF-Eval benchmark?

A challenging multimodal instruction-following benchmark that includes both compose-level constraints for output responses and perception-level constraints tied to input images, with comprehensive evaluation pipeline.

What is the MM IF-Eval leaderboard?

The MM IF-Eval leaderboard ranks 1 AI models based on their performance on this benchmark. Currently, Pixtral-12B by Mistral AI leads with a score of 0.527. The average score across all models is 0.527.

What is the highest MM IF-Eval score?

The highest MM IF-Eval score is 0.527, achieved by Pixtral-12B from Mistral AI.

How many models are evaluated on MM IF-Eval?

1 models have been evaluated on the MM IF-Eval benchmark, with 0 verified results and 1 self-reported results.

Where can I find the MM IF-Eval paper?

The MM IF-Eval paper is available at https://arxiv.org/abs/2504.07957. The paper details the methodology, dataset construction, and evaluation criteria.

What categories does MM IF-Eval cover?

MM IF-Eval is categorized under multimodal, reasoning, and structured output. The benchmark evaluates multimodal models.

What is the best open-source model on MM IF-Eval?

Pixtral-12B by Mistral AI is the top-ranked open-source model on MM IF-Eval, with a score of 0.527 (rank #1).

How recent are the MM IF-Eval leaderboard results?

The MM IF-Eval leaderboard was last updated in May 2026 and currently includes 1 evaluated models.

More evaluations to explore

Related benchmarks in the same category

View all multimodal
GPQA

A challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. Questions are Google-proof and extremely difficult, with PhD experts reaching 65% accuracy.

reasoning
216 models
MMLU-Pro

A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.

reasoning
120 models
AIME 2025

All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deductions and structured symbolic reasoning.

reasoning
109 models
MMLU

Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains

reasoning
99 models
SWE-Bench Verified

A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.

reasoning
92 models
Humanity's Last Exam

Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions

reasoningmultimodal
77 models