llmboard.aiAI model intelligence
Home

Model Rankings

OverallOpen ModelsAgentCodingReasoningMathKnowledgeInstruction FollowingTextVision
Image GenerationImage Editing
Video GenerationImage to VideoVideo Editing
Text to SpeechSpeech to Text
Embeddings

Efficiency

Chat Token PricingImage PricingVideo PricingAudio Pricing
Chat Speed & LatencyProvider Reliability

Benchmarks

GPQAMMLU-ProAIME 2025SWE-Bench VerifiedMMLUHumanity's Last ExamLiveCodeBenchMATHHumanEvalMMMU-Pro
All Benchmarks

Tools

Model DirectoryCompare Models

Scoring & Data

Scoring & Data
393 models668 benchmarks

Leaderboard Center

Overall RankingCodingCore BenchmarksPrice & ValueRuntime Performance

Modalities

All ModelsImage GenerationImage EditingVideo GenerationImage-to-VideoVideo EditingText-to-SpeechSpeech-to-TextEmbeddings

Data & Methods

Scoring MethodAll BenchmarksReasoningMath

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai

reasoning benchmark

Multi-IF Leaderboard

Multi-IF benchmarks LLMs on multi-turn and multilingual instruction following. It expands upon IFEval by incorporating multi-turn sequences and translating English prompts into 7 other languages, resulting in 4,501 multilingual conversations with three turns each. The benchmark reveals that current leading LLMs struggle with maintaining accuracy in multi-turn instructions and shows higher error rates for non-Latin script languages.

Updated Aug 17, 2026

Models23
Model coverage23
MetricScore
EvidenceB

On this page

  • Ranking
  • Highlights
  • Distribution
  • Top models
  • About
  • FAQ

Multi-IF Ranking

Higher score ranks better on this benchmark.

23 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelACQwen3-235B-A22B-Thinking-2507Alibaba Cloud / Qwen TeamScore80.6%Percentile100.0%Participants23EvidenceCEvaluatedAug 17, 2026
Rank02ModelLALFM2.5-2.6BLiquid AIScore80.1%Percentile95.5%Participants23EvidenceCEvaluatedAug 17, 2026
Rank03ModelOPo3-miniOpenAIScore79.5%Percentile90.9%Participants23EvidenceCEvaluatedAug 17, 2026
Rank04ModelACQwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen TeamScore79.1%Percentile86.4%Participants23EvidenceCEvaluatedAug 17, 2026
Rank05ModelACQwen3 VL 32B ThinkingAlibaba Cloud / Qwen TeamScore78.0%Percentile81.8%Participants23EvidenceCEvaluatedAug 17, 2026
Rank06ModelACQwen3-Next-80B-A3B-ThinkingAlibaba Cloud / Qwen TeamScore77.8%Percentile77.3%Participants23EvidenceCEvaluatedAug 17, 2026
Rank07ModelACQwen3-235B-A22B-Instruct-2507Alibaba Cloud / Qwen TeamScore77.5%Percentile72.7%Participants23EvidenceCEvaluatedAug 17, 2026
Rank08ModelACQwen3 VL 235B A22B InstructAlibaba Cloud / Qwen TeamScore76.3%Percentile68.2%Participants23EvidenceCEvaluatedAug 17, 2026
Rank09ModelACQwen3-Next-80B-A3B-InstructAlibaba Cloud / Qwen TeamScore75.8%Percentile63.6%Participants23EvidenceCEvaluatedAug 17, 2026
Rank10ModelACQwen3 VL 8B InstructAlibaba Cloud / Qwen TeamScore75.1%Percentile59.1%Participants23EvidenceCEvaluatedAug 17, 2026
Rank11ModelACQwen3 VL 8B ThinkingAlibaba Cloud / Qwen TeamScore75.1%Percentile54.5%Participants23EvidenceCEvaluatedAug 17, 2026
Rank12ModelACQwen3 VL 4B ThinkingAlibaba Cloud / Qwen TeamScore73.6%Percentile50.0%Participants23EvidenceCEvaluatedAug 17, 2026
Rank13ModelACQwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen TeamScore73.0%Percentile45.5%Participants23EvidenceCEvaluatedAug 17, 2026
Rank14ModelACQwen3 30B A3BAlibaba Cloud / Qwen TeamScore72.2%Percentile40.9%Participants23EvidenceCEvaluatedAug 17, 2026
Rank15ModelACQwen3 VL 32B InstructAlibaba Cloud / Qwen TeamScore72.0%Percentile36.4%Participants23EvidenceCEvaluatedAug 17, 2026
Rank16ModelOPGPT-4.1OpenAIScore70.8%Percentile31.8%Participants23EvidenceCEvaluatedAug 17, 2026
Rank17ModelOPGPT-4.5OpenAIScore70.8%Percentile27.3%Participants23EvidenceCEvaluatedAug 17, 2026
Rank18ModelOPGPT-4.1 miniOpenAIScore67.0%Percentile22.7%Participants23EvidenceCEvaluatedAug 17, 2026
Rank19ModelACQwen3 VL 30B A3B InstructAlibaba Cloud / Qwen TeamScore66.1%Percentile18.2%Participants23EvidenceCEvaluatedAug 17, 2026
Rank20ModelOPGPT-4oOpenAIScore60.9%Percentile13.6%Participants23EvidenceCEvaluatedAug 17, 2026
Rank21ModelLALFM2.5-VL-3BLiquid AIScore59.4%Percentile9.1%Participants23EvidenceCEvaluatedAug 17, 2026
Rank22ModelOPGPT-4.1 nanoOpenAIScore57.2%Percentile4.5%Participants23EvidenceCEvaluatedAug 17, 2026
Rank23ModelCONorth Micro Vision InstructCohereScore37.3%Percentile0.0%Participants23EvidenceCEvaluatedAug 17, 2026

Multi-IF Highlights

The leading models and scores on this benchmark.

Rank #1Qwen3-235B-A22B-Thinking-250780.6%Rank #2LFM2.5-2.6B80.1%Rank #3o3-mini79.5%Rank #4Qwen3 VL 235B A22B Thinking79.1%

Multi-IF Score Distribution

A closer view of the leading scores on this benchmark.

Multi-IF

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.

Ranking basisThis multi-if AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    AC
    Qwen3-235B-A22B-Thinking-2507Alibaba Cloud / Qwen Team
    Score
    80.6%

    Strengths

    • Ranks #1 of 23 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Multi-IF, not total model capability
  2. 02
    LA
    LFM2.5-2.6BLiquid AI
    Score
    80.1%

    Strengths

    • Ranks #2 of 23 compared models
    • 95th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Multi-IF, not total model capability
  3. 03
    OP
    o3-miniOpenAI
    Score
    79.5%
    Price
    $1.1 input / $4.4 output per 1M tokens
    Speed
    Up to 115 tok/s via Azure

    Strengths

    • Ranks #3 of 23 compared models
    • 91th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Multi-IF, not total model capability
  4. 04
    AC
    Qwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen Team
    Score
    79.1%

    Strengths

    • Ranks #4 of 23 compared models
    • 86th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Multi-IF, not total model capability
  5. 05
    AC
    Qwen3 VL 32B ThinkingAlibaba Cloud / Qwen Team
    Score
    78.0%

    Strengths

    • Ranks #5 of 23 compared models
    • 82th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Multi-IF, not total model capability

Selection summary

Best AI Models for Multi-IF

Qwen3-235B-A22B-Thinking-2507 currently leads Multi-IF with 80.6%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.

Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.

Benchmark rank #1Qwen3-235B-A22B-Thinking-250780.6%Benchmark rank #2LFM2.5-2.6B80.1%Benchmark rank #3o3-mini79.5% · $1.1 input / $4.4 output per 1M tokens

What is Multi-IF?

What Multi-IF measures and how its scores work.

Multi-IF benchmarks LLMs on multi-turn and multilingual instruction following. It expands upon IFEval by incorporating multi-turn sequences and translating English prompts into 7 other languages, resulting in 4,501 multilingual conversations with three turns each. The benchmark reveals that current leading LLMs struggle with maintaining accuracy in multi-turn instructions and shows higher error rates for non-Latin script languages.

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
reasoning
Score direction
higher
LLMBoard eligible
Yes
Evaluation key
multi-if|llm-stats-current

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 80.6%.

What does Multi-IF measure?

Multi-IF benchmarks LLMs on multi-turn and multilingual instruction following. It expands upon IFEval by incorporating multi-turn sequences and translating English prompts into 7 other languages, resulting in 4,501 multilingual conversations with three turns each. The benchmark reveals that current leading LLMs struggle with maintaining accuracy in multi-turn instructions and shows higher error rates for non-Latin script languages.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

23 model results are currently shown.

Does this benchmark affect the overall score?

Yes. This benchmark can contribute to the current LLMBoard capability score.