instruction following benchmark
IFBench Leaderboard
Instruction Following Benchmark evaluating model's ability to follow complex instructions
Updated Aug 17, 2026
Models34
Model coverage34
MetricScore
EvidenceB
IFBench Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score82.8% | Percentile100.0% | Participants34 | EvidenceC | Evaluated |
| Rank02 | ModelTM | Score82.2% | Percentile97.0% | Participants34 | EvidenceC | Evaluated |
| Rank03 | ModelNV | Score81.7% | Percentile93.9% | Participants34 | EvidenceC | Evaluated |
| Rank04 | ModelNR | Score81.2% | Percentile90.9% | Participants34 | EvidenceC | Evaluated |
| Rank05 | ModelAM Nova 2 ProAmazon | Score80.2% | Percentile87.9% | Participants34 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score79.5% | Percentile84.8% | Participants34 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score79.1% | Percentile81.8% | Participants34 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score79.1% | Percentile78.8% | Participants34 | EvidenceC | Evaluated |
| Rank09 | ModelME | Score77.0% | Percentile75.8% | Participants34 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score76.5% | Percentile72.7% | Participants34 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score76.5% | Percentile69.7% | Participants34 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score76.1% | Percentile66.7% | Participants34 | EvidenceC | Evaluated |
| Rank13 | ModelMI | Score75.0% | Percentile63.6% | Participants34 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score74.2% | Percentile60.6% | Participants34 | EvidenceC | Evaluated |
| Rank15 | ModelCO | Score74.0% | Percentile57.6% | Participants34 | EvidenceC | Evaluated |
| Rank16 | ModelNV | Score72.6% | Percentile54.5% | Participants34 | EvidenceC | Evaluated |
| Rank17 | ModelNV | Score71.9% | Percentile51.5% | Participants34 | EvidenceC | Evaluated |
| Rank18 | ModelIN | Score71.0% | Percentile48.5% | Participants34 | EvidenceC | Evaluated |
| Rank19 | ModelAMNova 2 LiteAmazon | Score70.8% | Percentile45.5% | Participants34 | EvidenceC | Evaluated |
| Rank20 | ModelAC | Score70.2% | Percentile42.4% | Participants34 | EvidenceC | Evaluated |
| Rank21 | ModelMI | Score70.0% | Percentile39.4% | Participants34 | EvidenceC | Evaluated |
| Rank22 | ModelOP | Score69.5% | Percentile36.4% | Participants34 | EvidenceC | Evaluated |
| Rank23 | ModelMI | Score69.0% | Percentile33.3% | Participants34 | EvidenceC | Evaluated |
| Rank24 | ModelMA | Score69.0% | Percentile30.3% | Participants34 | EvidenceC | Evaluated |
| Rank25 | ModelAMNova 2 OmniAmazon | Score68.7% | Percentile27.3% | Participants34 | EvidenceC | Evaluated |
| Rank26 | ModelLA | Score67.3% | Percentile24.2% | Participants34 | EvidenceC | Evaluated |
| Rank27 | ModelAC | Score64.5% | Percentile21.2% | Participants34 | EvidenceC | Evaluated |
| Rank28 | ModelAC | Score59.2% | Percentile18.2% | Participants34 | EvidenceC | Evaluated |
| Rank29 | ModelLA | Score59.2% | Percentile15.2% | Participants34 | EvidenceC | Evaluated |
| Rank30 | ModelMA | Score48.0% | Percentile12.1% | Participants34 | EvidenceC | Evaluated |
| Rank31 | ModelAC | Score41.3% | Percentile9.1% | Participants34 | EvidenceC | Evaluated |
| Rank32 | ModelAMNova 2 SonicAmazon | Score37.5% | Percentile6.1% | Participants34 | EvidenceC | Evaluated |
| Rank33 | ModelLA | Score25.8% | Percentile3.0% | Participants34 | EvidenceC | Evaluated |
| Rank34 | ModelAC | Score21.0% | Percentile0.0% | Participants34 | EvidenceC | Evaluated |
IFBench Highlights
The leading models and scores on this benchmark.
IFBench Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for IFBench
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis ifbench 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.
- 01ACQwen3.8 MaxAlibaba Cloud / Qwen Team
- Score
- 82.8%
- Price
- $2.0 input / $6.0 output per 1M tokens
Strengths
- Ranks #1 of 34 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures IFBench, not total model capability
- 02TMInkling-SmallThinking Machines Lab
- Score
- 82.2%
Strengths
- Ranks #2 of 34 compared models
- 97th percentile on this benchmark
- C evidence result
Considerations
- This result measures IFBench, not total model capability
- 03NV
- Score
- 81.7%
- Price
- $0.50 input / $2.5 output per 1M tokens
Strengths
- Ranks #3 of 34 compared models
- 94th percentile on this benchmark
- C evidence result
Considerations
- This result measures IFBench, not total model capability
- 04NRHermes 3 70BNous Research
- Score
- 81.2%
Strengths
- Ranks #4 of 34 compared models
- 91th percentile on this benchmark
- C evidence result
Considerations
- This result measures IFBench, not total model capability
- 05AMNova 2 ProAmazon
- Score
- 80.2%
Strengths
- Ranks #5 of 34 compared models
- 88th percentile on this benchmark
- C evidence result
Considerations
- This result measures IFBench, not total model capability
Selection summary
Best AI Models for IFBench
Qwen3.8 Max currently leads IFBench with 82.8%. 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.
What is IFBench?
What IFBench measures and how its scores work.
Instruction Following Benchmark evaluating model's ability to follow complex instructions
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- IFBench
- Modality
- text
- Primary category
- instruction following
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- ifbench|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about IFBench.
Which model scores highest on IFBench?
Qwen3.8 Max is currently ranked first with 82.8%.
What does IFBench measure?
Instruction Following Benchmark evaluating model's ability to follow complex instructions
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
34 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.