chat benchmark
IFEval Leaderboard
IFEval is an Instruction-Following Evaluation benchmark for large language models that uses around 500 prompts with one or more verifiable constraints across 25 types of instructions.
Updated Sep 5, 2026
Models68
Model coverage68
MetricScore
EvidenceB
IFEval Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score95.00% | Percentile100.00% | Participants68 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score94.60% | Percentile98.51% | Participants68 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score94.30% | Percentile97.01% | Participants68 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score94.30% | Percentile95.52% | Participants68 | EvidenceC | Evaluated |
| Rank05 | ModelOP | Score93.90% | Percentile94.03% | Participants68 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score93.40% | Percentile92.54% | Participants68 | EvidenceC | Evaluated |
| Rank07 | ModelAN | Score93.20% | Percentile91.04% | Participants68 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score92.60% | Percentile89.55% | Participants68 | EvidenceC | Evaluated |
| Rank09 | ModelME | Score92.10% | Percentile88.06% | Participants68 | EvidenceC | Evaluated |
| Rank10 | ModelAM Nova ProAmazon | Score92.10% | Percentile86.57% | Participants68 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score91.90% | Percentile85.07% | Participants68 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score91.50% | Percentile83.58% | Participants68 | EvidenceC | Evaluated |
| Rank13 | ModelGO | Score90.40% | Percentile82.09% | Participants68 | EvidenceC | Evaluated |
| Rank14 | ModelNV | Score90.30% | Percentile80.60% | Participants68 | EvidenceC | Evaluated |
| Rank15 | ModelGO | Score90.20% | Percentile79.10% | Participants68 | EvidenceC | Evaluated |
| Rank16 | ModelMA | Score89.80% | Percentile77.61% | Participants68 | EvidenceC | Evaluated |
| Rank17 | ModelMA | Score89.80% | Percentile76.12% | Participants68 | EvidenceC | Evaluated |
| Rank18 | ModelAC | Score89.80% | Percentile74.63% | Participants68 | EvidenceC | Evaluated |
| Rank19 | ModelAMNova LiteAmazon | Score89.70% | Percentile73.13% | Participants68 | EvidenceC | Evaluated |
| Rank20 | ModelME | Score89.65% | Percentile71.64% | Participants68 | EvidenceC | Evaluated |
| Rank21 | ModelLA | Score89.60% | Percentile70.15% | Participants68 | EvidenceC | Evaluated |
| Rank22 | ModelNV | Score89.45% | Percentile68.66% | Participants68 | EvidenceC | Evaluated |
| Rank23 | ModelGO | Score88.90% | Percentile67.16% | Participants68 | EvidenceC | Evaluated |
| Rank24 | ModelAC | Score88.90% | Percentile65.67% | Participants68 | EvidenceC | Evaluated |
| Rank25 | ModelAC | Score88.70% | Percentile64.18% | Participants68 | EvidenceC | Evaluated |
| Rank26 | ModelME | Score88.60% | Percentile62.69% | Participants68 | EvidenceC | Evaluated |
| Rank27 | ModelOP | Score88.20% | Percentile61.19% | Participants68 | EvidenceC | Evaluated |
| Rank28 | ModelAC | Score88.20% | Percentile59.70% | Participants68 | EvidenceC | Evaluated |
| Rank29 | ModelAC | Score87.80% | Percentile58.21% | Participants68 | EvidenceC | Evaluated |
| Rank30 | ModelAC | Score87.80% | Percentile56.72% | Participants68 | EvidenceC | Evaluated |
IFEval Highlights
The leading models and scores on this benchmark.
IFEval Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for IFEval
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is IFEval?
What IFEval measures and how its scores work.
IFEval is an Instruction-Following Evaluation benchmark for large language models.
IFEval measures Score as a ratio for performance on prompts containing verifiable instructions and constraints.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- IFEval
- 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 IFEval.
Which model scores highest on IFEval?
Qwen3.5-27B is currently ranked first with 95.00%.
What are the top three models on IFEval?
The current leaders are Qwen3.5-27B (95.00%), Qwen3.7-Plus (94.60%), and Qwen3.7 Max (94.30%).
Which IFEval model has the lowest official input price?
Nova Micro has the lowest matched official input price at $0.04 input / $0.14 output per 1M tokens.
Which models are fastest among IFEval results?
The fastest matched records are Llama 3.3 70B Instruct (2,220.00 tok/s via Cerebras), Llama 3.1 8B Instruct (2,047.00 tok/s via Cerebras), and Llama 3.1 70B Instruct (1,204.00 tok/s via Cerebras).
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.