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 Directory

Scoring & Data

Scoring & Data
1367 models808 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 Benchmarks
llmboard.aiCopyright 2026 llmboard.ai

instruction language context benchmark

AA IFBench Leaderboard

294 prompts; loose prompt-level accuracy; 5 runs; pass@1

Updated Sep 22, 2026

Models100
Model coverage168
Metricifbench
EvidenceB

On this page

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

AA IFBench Ranking

Higher ifbench ranks better on this benchmark.

30 of 100 rows
Columns

Show columns

Sort by
Rank
Model
ifbench
Percentile
Participants
Evidence
Evaluated
Rank01ModelXAGrok 4.3xAIifbench83.33%Percentile100.00%Participants168EvidenceBEvaluatedN/A
Rank02ModelMIMiniMax M3MiniMaxifbench82.86%Percentile99.40%Participants168EvidenceBEvaluatedN/A
Rank03ModelNVNemotron 3 Ultra (550B A55B)NVIDIAifbench81.36%Percentile98.80%Participants168EvidenceBEvaluatedN/A
Rank04ModelACQwen3.7 MaxAlibaba Cloud / Qwen Teamifbench80.54%Percentile98.20%Participants168EvidenceBEvaluatedN/A
Rank05ModelXIMiMo-V2.5-ProXiaomiifbench79.86%Percentile97.60%Participants168EvidenceBEvaluatedN/A
Rank06ModelACQwen3.5-397B-A17BAlibaba Cloud / Qwen Teamifbench78.78%Percentile97.01%Participants168EvidenceBEvaluatedN/A
Rank07ModelGOGemini 3 FlashGoogleifbench77.96%Percentile96.41%Participants168EvidenceBEvaluatedN/A
Rank08ModelACQwen3.7-PlusAlibaba Cloud / Qwen Teamifbench77.96%Percentile95.81%Participants168EvidenceBEvaluatedN/A
Rank09ModelOPGPT-5.2 CodexOpenAIifbench77.62%Percentile95.21%Participants168EvidenceBEvaluatedN/A
Rank10ModelGOGemini 3.1 Flash-LiteGoogleifbench77.21%Percentile94.61%Participants168EvidenceBEvaluatedN/A
Rank11ModelGOGemini 3.1 ProGoogleifbench77.14%Percentile94.01%Participants168EvidenceBEvaluatedN/A
Rank12ModelGOGemini 3.5 FlashGoogleifbench76.33%Percentile93.41%Participants168EvidenceBEvaluatedN/A
Rank13ModelZAGLM-5.1Zhipu AIifbench76.26%Percentile92.81%Participants168EvidenceBEvaluatedN/A
Rank14ModelMAKimi K2.6Moonshot AIifbench75.99%Percentile92.22%Participants168EvidenceBEvaluatedN/A
Rank15ModelOPGPT-5.4 nanoOpenAIifbench75.92%Percentile91.62%Participants168EvidenceBEvaluatedN/A
Rank16ModelMEMuse SparkMetaifbench75.92%Percentile91.02%Participants168EvidenceBEvaluatedN/A
Rank17ModelOPGPT-5.5OpenAIifbench75.85%Percentile90.42%Participants168EvidenceBEvaluatedN/A
Rank18ModelACQwen3.5-122B-A10BAlibaba Cloud / Qwen Teamifbench75.71%Percentile89.82%Participants168EvidenceBEvaluatedN/A
Rank19ModelMIMiniMax M2.7MiniMaxifbench75.71%Percentile89.22%Participants168EvidenceBEvaluatedN/A
Rank20ModelACQwen3.5-27BAlibaba Cloud / Qwen Teamifbench75.58%Percentile88.62%Participants168EvidenceBEvaluatedN/A
Rank21ModelGOGemma 4 31BGoogleifbench75.58%Percentile88.02%Participants168EvidenceBEvaluatedN/A
Rank22ModelOPGPT-5 miniOpenAIifbench75.44%Percentile87.43%Participants168EvidenceBEvaluatedN/A
Rank23ModelOPGPT-5.3 CodexOpenAIifbench75.37%Percentile86.83%Participants168EvidenceBEvaluatedN/A
Rank24ModelACQwen3.6 PlusAlibaba Cloud / Qwen Teamifbench75.17%Percentile86.23%Participants168EvidenceBEvaluatedN/A
Rank25ModelOPGPT-5 CodexOpenAIifbench74.15%Percentile85.63%Participants168EvidenceBEvaluatedN/A
Rank26ModelCOCommand A+Cohereifbench73.95%Percentile85.03%Participants168EvidenceBEvaluatedN/A
Rank27ModelOPGPT-5.4OpenAIifbench73.95%Percentile84.43%Participants168EvidenceBEvaluatedN/A
Rank28ModelGOGemma 4 12BGoogleifbench73.54%Percentile83.83%Participants168EvidenceBEvaluatedN/A
Rank29ModelZAGLM-5.2Zhipu AIifbench73.33%Percentile83.23%Participants168EvidenceBEvaluatedN/A
Rank30ModelOPGPT-5.4 miniOpenAIifbench73.27%Percentile82.63%Participants168EvidenceBEvaluatedN/A

AA IFBench Highlights

The leading models and scores on this benchmark.

AA IFBench Score Distribution

A closer view of the leading scores on this benchmark.

AA IFBench

The Top AI Models for AA IFBench

The first five results on this benchmark, with official price and output speed added where the model identity can be matched.

What is AA IFBench?

What AA IFBench measures and how its scores work.

294 prompts; loose prompt-level accuracy; 5 runs; pass@1

Scores are shown in ratio. This benchmark is verified and has an evidence level of B.

Family
ifbench
Modality
text
Primary category
instruction language context
Score direction
higher
LLMBoard eligible
Yes
Evaluation key
overall

Artificial Analysis. Benchmark scores retain their original unit. Overall score eligibility is shown separately.

FAQ

Common questions about AA IFBench.

Which model scores highest on AA IFBench?

Grok 4.3 is currently ranked first with 83.33%.

What are the top three models on AA IFBench?

The current leaders are Grok 4.3 (83.33%), MiniMax M3 (82.86%), and Nemotron 3 Ultra (550B A55B) (81.36%).

Which AA IFBench model has the lowest official input price?

GPT-5 nano has the lowest matched official input price at $0.05 input / $0.40 output per 1M tokens.

Which models are fastest among AA IFBench results?

The fastest matched records are Mercury 2 (1,009.00 tok/s via Inception), GPT OSS 20B (1,000.00 tok/s via Groq), and GPT OSS 120B (500.00 tok/s via Groq).

Does the highest ifbench 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.

What does AA IFBench measure?

294 prompts; loose prompt-level accuracy; 5 runs; pass@1

Is a higher ifbench better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

100 model results are currently shown.

Does this benchmark affect the overall score?

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

Rank #1Grok 4.383.33%
Rank #2MiniMax M382.86%
Rank #3Nemotron 3 Ultra (550B A55B)81.36%
Rank #4Qwen3.7 Max80.54%

Ranking basisThis aa ifbench AI model leaderboard uses descending ifbench in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    XA
    Grok 4.3xAI
    ifbench
    83.33%
    Price
    $1.3 input / $2.5 output per 1M tokens

    Strengths

    • Ranks #1 of 168 compared models
    • 100th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures AA IFBench, not total model capability
  2. 02
    MI
    MiniMax M3MiniMax
    ifbench
    82.86%
    Price
    $0.30 input / $1.2 output per 1M tokens
    Speed
    Up to 45.04 tok/s via MiniMax

    Strengths

    • Ranks #2 of 168 compared models
    • 99th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures AA IFBench, not total model capability
  3. 03
    NV
    NVIDIA
    ifbench
    81.36%

    Strengths

    • Ranks #3 of 168 compared models
    • 99th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures AA IFBench, not total model capability
  4. 04
    AC
    Alibaba Cloud / Qwen Team
    ifbench
    80.54%
    Price
    $2.5 input / $7.5 output per 1M tokens

    Strengths

    • Ranks #4 of 168 compared models
    • 98th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures AA IFBench, not total model capability
  5. 05
    XI
    Xiaomi
    ifbench
    79.86%
    Price
    $0.44 input / $0.87 output per 1M tokens

    Strengths

    • Ranks #5 of 168 compared models
    • 98th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures AA IFBench, not total model capability

Selection summary

Best AI Models for AA IFBench

Grok 4.3 currently leads AA IFBench with 83.33%. 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.

Nemotron 3 Ultra (550B A55B)
Qwen3.7 Max
MiMo-V2.5-Pro
Benchmark rank #1Grok 4.383.33% · $1.3 input / $2.5 output per 1M tokens
Benchmark rank #2MiniMax M382.86% · $0.30 input / $1.2 output per 1M tokens
Benchmark rank #3Nemotron 3 Ultra (550B A55B)81.36%