Instruction Following benchmark

# IFEval leaderboard

> IFEval results for 57 AI models, led by Grok-3 mini at 95.1%. What the benchmark measures, who runs it, and a source for every score.
- Canonical page: https://noometry.com/benchmarks/ifeval
- Last updated: 2026-10-10
- Title: IFEval Leaderboard (October 2026): Scores by Model

As of October 2026, Grok-3 mini has the highest published IFEval score on Noometry at 95.1%, out of 57 models with results.

Last verified October 10, 2026

## About IFEval

Prompts with checkable instructions such as word counts, required keywords or output format, scored by strict automatic checks. HELM Capabilities run.

- **Category:** [Instruction Following](https://noometry.com/best/instruction-following)
- **Introduced:** 2023
- **Size:** 541 prompts
- **Format:** Verifiable constraints
- **Unit:** Percent (random guessing ≈ 0%)
- **Official site:** [crfm.stanford.edu](https://crfm.stanford.edu/helm/capabilities/latest/)

## Top 15 models

Top models on IFEval

1.  Grok-3 mini 95.1%
2.  Grok 4 94.9%
3.  GPT-5.1 93.5%
4.  GPT-5 Nano 93.2%
5.  o4-mini 92.8%
6.  GPT-5 Mini 92.7%
7.  Claude Opus 4 91.8%
8.  Llama 4 Maverick 90.8%
9.  GPT-4.1 mini 90.4%
10.  Gemini 2.5 Flash 89.8%
11.  Granite 4.0 H Small 89%
12.  Grok 3 88.4%
13.  Gemini 3 Pro 87.7%
14.  Mistral Large 87.7%
15.  GPT-5 87.5%
16.  868890929496

Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise)

## All results

IFEval results by model
| # | Model | Provider | Score | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [Grok-3 mini](https://noometry.com/models/grok-3-mini) | [xAI](https://noometry.com/providers/xai) | 95.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 2 | [Grok 4](https://noometry.com/models/grok-4) | [xAI](https://noometry.com/providers/xai) | 94.9% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 3 | [GPT-5.1](https://noometry.com/models/gpt-5-1) | [OpenAI](https://noometry.com/providers/openai) | 93.5% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 4 | [GPT-5 Nano](https://noometry.com/models/gpt-5-nano) | [OpenAI](https://noometry.com/providers/openai) | 93.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 5 | [o4-mini](https://noometry.com/models/o4-mini) | [OpenAI](https://noometry.com/providers/openai) | 92.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 6 | [GPT-5 Mini](https://noometry.com/models/gpt-5-mini) | [OpenAI](https://noometry.com/providers/openai) | 92.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 7 | [Claude Opus 4](https://noometry.com/models/claude-opus-4) | [Anthropic](https://noometry.com/providers/anthropic) | 91.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 8 | [Llama 4 Maverick](https://noometry.com/models/llama-4-maverick) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 90.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 9 | [GPT-4.1 mini](https://noometry.com/models/gpt-4-1-mini) | [OpenAI](https://noometry.com/providers/openai) | 90.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 10 | [Gemini 2.5 Flash](https://noometry.com/models/gemini-2-5-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 89.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 11 | [Granite 4.0 H Small](https://noometry.com/models/ibm-granite-h-small) | [IBM](https://noometry.com/providers/ibm) | 89% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 12 | [Grok 3](https://noometry.com/models/grok-3) | [xAI](https://noometry.com/providers/xai) | 88.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 13 | [Gemini 3 Pro](https://noometry.com/models/gemini-3-pro) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 87.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 14 | [Mistral Large](https://noometry.com/models/mistral-large) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 87.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 15 | [GPT-5](https://noometry.com/models/gpt-5) | [OpenAI](https://noometry.com/providers/openai) | 87.5% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 16 | [o3](https://noometry.com/models/o3) | [OpenAI](https://noometry.com/providers/openai) | 86.9% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 17 | [Claude 3.5 Sonnet](https://noometry.com/models/claude-3-5-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 85.5% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 18 | [Kimi K2 (Jul 2025)](https://noometry.com/models/kimi-k2) | [Moonshot AI](https://noometry.com/providers/moonshot) | 85% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 19 | [Claude Sonnet 4.5](https://noometry.com/models/claude-sonnet-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 85% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 20 | [Granite 4.0 Micro](https://noometry.com/models/granite-4-0-micro) | [IBM](https://noometry.com/providers/ibm) | 84.9% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 21 | [GPT-4.1 nano](https://noometry.com/models/gpt-4-1-nano) | [OpenAI](https://noometry.com/providers/openai) | 84.3% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 22 | [Gemini 2.0 Flash (Feb 2025)](https://noometry.com/models/gemini-2-0-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 84.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 23 | [Claude Sonnet 4](https://noometry.com/models/claude-sonnet-4) | [Anthropic](https://noometry.com/providers/anthropic) | 84% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 24 | [Gemini 2.5 Pro](https://noometry.com/models/gemini-2-5-pro) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 84% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 25 | [GPT-4.1](https://noometry.com/models/gpt-4-1) | [OpenAI](https://noometry.com/providers/openai) | 83.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 26 | [Gemini 1.5 Pro (May 2024)](https://noometry.com/models/gemini-1-5-pro) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 83.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 27 | [gpt-oss-120b](https://noometry.com/models/gpt-oss-120b) | [OpenAI](https://noometry.com/providers/openai) | 83.6% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 28 | [Qwen3 235B-A22B](https://noometry.com/models/qwen3-235b-a22b) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 83.5% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 29 | [Claude 3.7 Sonnet](https://noometry.com/models/claude-3-7-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 83.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 30 | [DeepSeek-V3](https://noometry.com/models/deepseek-v3) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 83.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 31 | [Gemini 1.5 Flash (May 2024)](https://noometry.com/models/gemini-1-5-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 83.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 32 | [Gemini 2.0 Flash-Lite](https://noometry.com/models/gemini-2-0-flash-lite) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 82.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 33 | [Llama 3.1-70B](https://noometry.com/models/llama-3-1-70b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 82.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 34 | [Llama 4 Scout](https://noometry.com/models/llama-4-scout) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 81.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 35 | [GPT-4o](https://noometry.com/models/gpt-4o) | [OpenAI](https://noometry.com/providers/openai) | 81.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 36 | [Amazon Nova Pro](https://noometry.com/models/amazon-nova-pro) | [Amazon](https://noometry.com/providers/amazon) | 81.5% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 37 | [GLM-4.5-Air](https://noometry.com/models/glm-4-5-air) | [Z.ai (Zhipu)](https://noometry.com/providers/zai) | 81.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 38 | [Llama 3.1-405B](https://noometry.com/models/llama-3-1-405b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 81.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 39 | [Gemini 2.5 Flash-Lite](https://noometry.com/models/gemini-2-5-flash-lite) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 81% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 40 | [Qwen3-Next 80B-A3B Instruct](https://noometry.com/models/qwen3-next-80b-a3b-instruct) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 81% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 41 | [Qwen2.5 72B Instruct](https://noometry.com/models/qwen2-5-72b-instruct) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 80.6% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 42 | [Nova Premier 1.0](https://noometry.com/models/nova-premier-1-0) | [Amazon](https://noometry.com/providers/amazon) | 80.3% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 43 | [Claude Haiku 4.5](https://noometry.com/models/claude-haiku-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 80.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 44 | [Claude 3.5 Haiku](https://noometry.com/models/claude-3-5-haiku) | [Anthropic](https://noometry.com/providers/anthropic) | 79.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 45 | [DeepSeek-R1](https://noometry.com/models/deepseek-r1) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 78.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 46 | [GPT-4o mini](https://noometry.com/models/gpt-4o-mini) | [OpenAI](https://noometry.com/providers/openai) | 78.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 47 | [Olmo 2 0325 32b Instruct](https://noometry.com/models/olmo-2-0325-32b-instruct) |  [![](/logos/ai2.svg) Allen Institute for AI (Ai2)](https://noometry.com/providers/ai2) | 78% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 48 | [Amazon Nova Lite](https://noometry.com/models/amazon-nova-lite) | [Amazon](https://noometry.com/providers/amazon) | 77.6% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 49 | [Amazon Nova Micro](https://noometry.com/models/amazon-nova-micro) | [Amazon](https://noometry.com/providers/amazon) | 76% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 50 | [Mistral Small 3.1](https://noometry.com/models/mistral-small-3-1) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 75% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 51 | [Llama 3.1-8B](https://noometry.com/models/llama-3-1-8b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 74.3% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 52 | [Qwen2.5 7B Instruct](https://noometry.com/models/qwen2-5-7b-instruct) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 74.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 53 | [gpt-oss-20b](https://noometry.com/models/gpt-oss-20b) | [OpenAI](https://noometry.com/providers/openai) | 73.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 54 | [OLMo 2 Furious 13B](https://noometry.com/models/olmo-2-furious-13b) |  [![](/logos/ai2.svg) Allen Institute for AI (Ai2)](https://noometry.com/providers/ai2) | 73% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 55 | [Mixtral 8x22B](https://noometry.com/models/mixtral-8x22b) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 72.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 56 | [Mixtral 8x7B](https://noometry.com/models/mixtral-8x7b) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 57.5% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 57 | [Mistral](https://noometry.com/models/mistral) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 56.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |

## Compare the leaders

-   [Grok-3 mini vs Grok 4](https://noometry.com/compare/grok-3-mini-vs-grok-4)
-   [Grok-3 mini vs GPT-5.1](https://noometry.com/compare/gpt-5-1-vs-grok-3-mini)
-   [Grok-3 mini vs GPT-5 Nano](https://noometry.com/compare/gpt-5-nano-vs-grok-3-mini)
-   [Grok-3 mini vs o4-mini](https://noometry.com/compare/grok-3-mini-vs-o4-mini)
-   [Grok 4 vs GPT-5.1](https://noometry.com/compare/gpt-5-1-vs-grok-4)
-   [Grok 4 vs GPT-5 Nano](https://noometry.com/compare/gpt-5-nano-vs-grok-4)

## Other instruction following benchmarks

-   [LiveBench Instruction Following](https://noometry.com/benchmarks/livebench-if)
-   [LMArena Instruction Following](https://noometry.com/benchmarks/arena-instruction-following)

## Frequently asked questions

### What does IFEval measure?

Prompts with checkable instructions such as word counts, required keywords or output format, scored by strict automatic checks. HELM Capabilities run.

### Which model has the highest IFEval score?

As of October 2026, Grok-3 mini has the highest published IFEval score on Noometry at 95.1%, out of 57 models with results.

### What is the best open-weight model on IFEval?

Llama 4 Maverick has the highest IFEval accuracy among open-weight models at 90.8%, ranking 8 of 57 overall.

### Cite this page

Noometry. (2026). IFEval leaderboard. Retrieved October 10, 2026, from https://noometry.com/benchmarks/ifeval

Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/benchmarks/ifeval.md).