chat benchmark
Multi-IF Leaderboard
Multi-IF benchmarks LLMs on multi-turn and multilingual instruction following using 4,501 three-turn conversations created by translating English prompts into 7 other languages.
Updated Sep 5, 2026
Models23
Model coverage23
MetricScore
EvidenceB
Multi-IF Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score80.60% | Percentile100.00% | Participants23 | EvidenceC | Evaluated |
| Rank02 | ModelLA | Score80.07% | Percentile95.45% | Participants23 | EvidenceC | Evaluated |
| Rank03 | ModelOP | Score79.50% | Percentile90.91% | Participants23 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score79.10% | Percentile86.36% | Participants23 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score78.00% | Percentile81.82% | Participants23 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score77.80% | Percentile77.27% | Participants23 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score77.50% | Percentile72.73% | Participants23 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score76.30% | Percentile68.18% | Participants23 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score75.80% | Percentile63.64% | Participants23 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score75.10% | Percentile59.09% | Participants23 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score75.10% | Percentile54.55% | Participants23 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score73.60% | Percentile50.00% | Participants23 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score73.00% | Percentile45.45% | Participants23 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score72.20% | Percentile40.91% | Participants23 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score72.00% | Percentile36.36% | Participants23 | EvidenceC | Evaluated |
| Rank16 | ModelOP | Score70.80% | Percentile31.82% | Participants23 | EvidenceC | Evaluated |
| Rank17 | ModelOP | Score70.80% | Percentile27.27% | Participants23 | EvidenceC | Evaluated |
| Rank18 | ModelOP | Score67.00% | Percentile22.73% | Participants23 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score66.10% | Percentile18.18% | Participants23 | EvidenceC | Evaluated |
| Rank20 | ModelOP | Score60.90% | Percentile13.64% | Participants23 | EvidenceC | Evaluated |
| Rank21 | ModelLA | Score59.40% | Percentile9.09% | Participants23 | EvidenceC | Evaluated |
| Rank22 | ModelOP | Score57.20% | Percentile4.55% | Participants23 | EvidenceC | Evaluated |
| Rank23 | ModelCO | Score37.30% | Percentile0.00% | Participants23 | EvidenceC | Evaluated |
Multi-IF Highlights
The leading models and scores on this benchmark.
Multi-IF Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for Multi-IF
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is Multi-IF?
What Multi-IF measures and how its scores work.
Multi-IF is a chat benchmark that expands upon IFEval with multi-turn sequences and multilingual conversations in English and 7 other languages.
It measures multi-turn and multilingual instruction-following performance with Score, reported as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- Multi-IF
- Modality
- text
- Primary category
- chat
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about Multi-IF.
Which model scores highest on Multi-IF?
Qwen3-235B-A22B-Thinking-2507 is currently ranked first with 80.60%.
What are the top three models on Multi-IF?
The current leaders are Qwen3-235B-A22B-Thinking-2507 (80.60%), LFM2.5-2.6B (80.07%), and o3-mini (79.50%).
Which Multi-IF model has the lowest official input price?
GPT-4.1 nano has the lowest matched official input price at $0.10 input / $0.40 output per 1M tokens.
Which models are fastest among Multi-IF results?
The fastest matched records are GPT-4.1 mini (467.39 tok/s via OpenAI), GPT-4.1 nano (138.14 tok/s via OpenAI), and GPT-4o (132.00 tok/s via OpenAI).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.