Neura Intelligence Index

AI Model Benchmarks

Q: What is the Neura Intelligence Index?

The Neura Intelligence Index is a composite score (0-100) that aggregates AI model performance across 15+ benchmarks spanning coding, math, reasoning, general knowledge, language, multimodal, safety, and agentic tasks. It uses min-max normalization and weighted category averaging to produce a single comparable score.

Q: How often are benchmark scores updated?

Benchmark scores are synced daily via an automated pipeline that aggregates data from official model cards, Papers With Code, HuggingFace Open LLM Leaderboard, LiveBench, and LMSYS Chatbot Arena.

Q: What does the confidence level mean?

Confidence reflects how many benchmarks a model has been tested on relative to the total. High means >70% benchmark coverage, medium is 40-70%, and low is <40%. Models with low coverage may have composite scores that shift as more benchmarks are added.

Q: How are category weights determined?

Category weights reflect real-world demand: Coding and Reasoning each get 20%, General and Math each get 15%, Language and Multimodal each get 10%, and Safety and Agent each get 5%. Weights are renormalized across available categories so models are not penalized for missing data.

Q: Can I compare specific models side by side?

Yes. Click any model to see its full benchmark profile, or use the comparison feature to compare up to 4 models side by side with radar charts and benchmark-by-benchmark scoring.

Compare AI model performance across 48+ benchmarks. Our composite index aggregates coding, math, reasoning, and language scores into a single intelligence ranking.

396

Models Ranked

Benchmarks

Aug 14, 2026

Last Updated

All(396)LLMs(374)Image(10)Video(8)Agent(4)

Data API Deep Dives

Highlights

At-a-glance rankings across the three dimensions that matter most.

Intelligence

Neura Intelligence Index; Higher is better

Context Window

Max input tokens; Higher is better

Price

USD per 1M output tokens; Lower is better

Leaderboard

Ranked by the Neura Intelligence Index — a weighted composite of 48 benchmarks across 8 categories.

Intelligence Tradeoffs

Find the sweet spot — models in the top-left quadrant offer the best value.

Intelligence vs. Output Cost ($/1M tokens)

xAI

Anthropic

DeepSeek

Google

OpenAI

ByteDance

Zhipu AI

Alibaba Cloud / Qwen Team

Microsoft

Mistral

Category Leaders

Top performers in each benchmark category.

Top Models per Task

Who leads on each individual benchmark — click any card to see full results.

Best in GPQA

GPQA Diamond (%)

Best in SWE-bench

SWE-bench Verified (%)

Best in MATH

MATH (%)

Best in MMLU-Pro

MMLU-Pro (%)

Best in HumanEval

HumanEval (%)

Best in Chatbot Arena Elo

Chatbot Arena Elo (Elo)

Best Model For Your Use Case

Different tasks need different strengths. These indices re-weight our benchmarks for specific workflows.

Coding & Engineering

Best for software development, code generation, and debugging

Claude Fable 597.5 2

Claude Opus 596.7 3

Gemini 3.7 Flash96.6 4

GLM-5.395.7 5

DeepSeek-V4-Pro-081394.3

Research & Analysis

Best for scientific research, data analysis, and complex reasoning

Claude Opus 596.7 2

Claude Fable 596.6 3

Gemini 3.7 Flash96.6 4

GLM-5.395.7 5

Muse Spark 1.195.5

Content & Copywriting

Best for writing, editing, summarization, and creative tasks

Claude Opus 596.7 2

Claude Fable 596.6 3

Gemini 3.7 Flash96.6 4

GLM-5.395.7 5

Muse Spark 1.195.5

Best Value

Highest intelligence per dollar — the cost-efficiency sweet spot

DeepSeek-V4-Flash-0731522.6 2

Step-3.5-Flash182.3 3

Gemma 4 31B177.3 4

DeepSeek-V4-Pro-0813109.0 5

GPT-5.6 Luna63.5

Methodology

The Neura Intelligence Index is a composite score (0-100) computed from 48+ individual benchmarks spanning 8 categories: coding, math, reasoning, general knowledge, language, multimodal, safety, and agentic tasks.

How it works

Normalize — Each benchmark score is min-max normalized to 0-1 across all models.
Categorize — Normalized scores are averaged within their benchmark category.
Weight — Category averages are combined using weights: Coding (20%), Reasoning (20%), General (15%), Math (15%), Language (10%), Multimodal (10%), Safety (5%), Agent (5%).
Scale — The weighted average is scaled to 0-100.

Coverage & confidence

Not all models have scores on all benchmarks. The confidence indicator reflects benchmark coverage: high (>70% of benchmarks), medium (40-70%), or low (<40%). Weights are renormalized across available categories so models aren't penalized for missing data.

Data sources

Scores are aggregated from official model cards, Papers With Code, HuggingFace Open LLM Leaderboard, LiveBench, and LMSYS Chatbot Arena. Each score includes a verification status (official, self-reported, or aggregated).

Frequently Asked Questions

What is the Neura Intelligence Index?▾

The Neura Intelligence Index is a composite score (0-100) that aggregates AI model performance across 15+ benchmarks spanning coding, math, reasoning, general knowledge, language, multimodal, safety, and agentic tasks. It uses min-max normalization and weighted category averaging to produce a single comparable score.

How often are benchmark scores updated?▾

Benchmark scores are synced daily via an automated pipeline that aggregates data from official model cards, Papers With Code, HuggingFace Open LLM Leaderboard, LiveBench, and LMSYS Chatbot Arena.

What does the confidence level mean?▾

Confidence reflects how many benchmarks a model has been tested on relative to the total. High means >70% benchmark coverage, medium is 40-70%, and low is <40%. Models with low coverage may have composite scores that shift as more benchmarks are added.

How are category weights determined?▾

Category weights reflect real-world demand: Coding and Reasoning each get 20%, General and Math each get 15%, Language and Multimodal each get 10%, and Safety and Agent each get 5%. Weights are renormalized across available categories so models are not penalized for missing data.

Can I compare specific models side by side?

AI Model Benchmarks

AI Model Benchmarks

Highlights

Intelligence

Context Window

Price

Leaderboard

Intelligence Tradeoffs

Intelligence vs. Output Cost ($/1M tokens)

Category Leaders

General Knowledge

Coding

Mathematics

Reasoning

Language

Multimodal

Safety

Agentic

Top Models per Task

Best in GPQA

Best in SWE-bench

Best in MATH

Best in MMLU-Pro

Best in HumanEval

Best in Chatbot Arena Elo

Best Model For Your Use Case

Coding & Engineering

Research & Analysis

Content & Copywriting

Best Value

Methodology

How it works

Coverage & confidence

Data sources

Frequently Asked Questions

Intelligence vs. Context Window

#	Model											Coverage
1—	X Grok 4.6 xAI🇺🇸2d	99.3	—	—	—	—	—	—	$6	—	—	low
2—	GPT-Image-1 OpenAI🇺🇸1y	98.7	—	—	—	—	—	—	—	—	—	low
3—	Claude Fable 5 Anthropic🇺🇸2mo	97.4	98.1	—	96.6	—	—	—	$50	—	—	low
4—	Z GLM-5.3 Zhipu AI🇨🇳0dOSS	95.8	—	—	95.7	—	—	—	—	—	—	low
5—	Claude Opus 5 Anthropic🇺🇸3w	95.5	—	—	96.7	—	—	—	$25	—	—	low
6—	DeepSeek-V4-Pro-0813 DeepSeek🇨🇳1dOSS	94.8	—	—	94.3	—	—	—	$0.87	—	—	low
7—	DeepSeek-V4-Flash-0731 DeepSeek🇨🇳2wOSS	94.1	—	—	—	—	—	—	$0.18	—	—	low
8—	Gemini 3.7 Flash Google🇺🇸1d	92.1	—	—	—	96.6	—	85.3	$4	—	—	low
9—	Claude Mythos Preview Anthropic🇺🇸	91.9	97.8	—	94.1	83.3	—	91.3	—	—	—	medium
10—	Claude Opus 4.8 Anthropic🇺🇸2mo	91.3	93.5	—	91.8	—	—	91.0	$25	—	—	high
11—	GPT-5.6 Sol OpenAI🇺🇸1mo	90.2	83.7	98.2	91.5	94.8	—	86	$30	—	—	medium
12—	Q Qwen3.8 Max Alibaba Cloud / Qwen Team🇨🇳1wOSS	89.9	92.5	—	82.5	95.3	—	88.4	—	—	—	medium
13—	Sora OpenAI🇺🇸1y	89.4	—	—	—	—	—	—	—	—	—	low
14—	M Kimi K3 Moonshot AI🇨🇳4wOSS	88.8	—	—	91.0	—	—	86.4	$15	—	—	high
15—	Veo 2 Google🇺🇸1y	88.2	—	—	—	—	—	—	—	—	—	low
16—	GPT-5.6 Terra OpenAI🇺🇸1mo	87.4	78.9	97.4	90.2	94.0	—	82.0	$12	—	—	medium
17—	X Grok 4.5 xAI🇺🇸4w	87.2	84.1	—	90.2	—	—	—	$6	—	—	low
18—	Claude Opus 4.7 Anthropic🇺🇸4mo	85.8	86.7	—	90.7	80.1	—	88.7	$25	—	—	high
19—	Claude Sonnet 5 Anthropic🇺🇸1mo	85.8	83.2	—	92.5	—	—	84.7	$10	—	—	medium
20—	X Grok-4 Heavy xAI🇺🇸	85.0	—	84.2	85.7	—	—	—	—	—	—	low
21—	GPT-5.1 High OpenAI🇺🇸9mo	84.6	—	83.8	85.3	—	—	—	—	—	—	low
22—	M Muse Spark 1.1 Meta🇺🇸1mo	84.3	69.9	—	95.5	—	—	84.8	$4	—	—	medium
23—	GPT-5.1 Medium OpenAI🇺🇸9mo	82.3	—	82.3	—	—	—	—	$10	—	—	low
24—	GPT-5 High OpenAI🇺🇸1y	81.3	—	77.3	84.4	—	—	—	—	—	—	low
25—	B Seed 2.1 Pro ByteDance🇨🇳1mo	81.0	71.9	—	91.0	—	—	83.4	—	—	—	high
26—	DALL-E 3 OpenAI🇺🇸2y	81.0	—	—	—	—	—	—	—	—	—	low
27—	Claude Opus 4.6 Anthropic🇺🇸6mo	80.9	82.2	84.0	86.7	82.1	—	67.8	$25	—	—	high
28—	B Seed 2.0 Pro ByteDance🇨🇳6mo	80.3	74.6	82.2	86.2	—	—	—	$3	—	—	medium
29—	GPT-5.1 Codex High OpenAI🇺🇸9mo	80.2	—	80.2	—	—	—	—	—	—	—	low
30—	ChatGPT-4o Latest OpenAI🇺🇸2y	80.1	—	—	80.1	—	—	—	—	—	—	low
31—	Claude Sonnet 4.5 Anthropic🇺🇸10mo	79.0	92.8	65.0	79.2	72.5	—	—	$15	—	—	medium
32—	GPT-5.5 Pro OpenAI🇺🇸3mo	78.9	—	57.5	92.3	—	—	—	—	—	—	low
33—	Claude Code Anthropic🇺🇸1y	78.6	66.7	—	—	—	—	—	—	—	—	low
34—	B Seed 2.1 Turbo ByteDance🇨🇳1mo	78.5	69.3	—	90.0	—	—	80.2	—	—	—	high
35—	Z GLM-5.2 Zhipu AI🇨🇳1moOSS	78.3	72.9	—	89.3	—	—	—	$3	—	—	medium
36—	X Grok-3 xAI🇺🇸1y	78.2	—	75.3	80.9	—	—	77.0	$15	—	—	low
37—	GPT-5 Medium OpenAI🇺🇸1y	78.0	—	68.3	85.3	—	—	—	—	—	—	low
38—	Q Qwen3.7 Max Alibaba Cloud / Qwen Team🇨🇳2mo	77.5	77.5	—	80.3	76.5	—	—	$4	—	—	medium
39—	X MiMo-V2-Pro Xiaomi🇨🇳4mo	77.4	77.4	—	—	—	—	—	—	—	—	low
40—	Z GLM-5V-Turbo Zhipu AI🇨🇳4mo	77.3	—	—	—	—	—	—	—	—	—	low
41—	GPT-5.2 Pro OpenAI🇺🇸8mo	77.0	—	84.2	72.1	—	—	—	—	—	—	medium
42—	ERNIE 5.0 Baidu🇨🇳6mo	76.8	—	65.0	73.9	92.5	—	—	—	—	—	medium
43—	X Grok-3 Mini xAI🇺🇸1y	76.4	—	71.5	80.1	—	—	—	—	—	—	low
44—	GPT-5.6 Luna OpenAI🇺🇸1mo	76.2	75.8	95.5	89.6	40.9	—	77.4	$1	—	—	medium
45—	GPT-5.1 OpenAI🇺🇸9mo	76.0	74.2	57.1	85.3	—	—	89.6	$10	—	—	medium
46—	GPT-5.1 Instant OpenAI🇺🇸9mo	76.0	74.2	57.1	85.3	—	—	89.6	$10	—	—	medium
47—	GPT-5.1 Thinking OpenAI🇺🇸9mo	76.0	74.2	57.1	85.3	—	—	89.6	—	—	—	medium
48—	Claude Sonnet 4.6 Anthropic🇺🇸5mo	76.0	80.2	—	80.4	73.2	—	71.0	$15	—	—	high
49—	o1 OpenAI🇺🇸1y	75.4	48.6	94.2	79.9	86.7	—	—	$60	—	—	high
50—	Claude Opus 4 Anthropic🇺🇸1y	75.2	58.2	75.5	79.9	96.0	—	68.1	$75	—	—	high
51—	B Seed 2.0 Lite ByteDance🇨🇳6mo	75.0	68.4	74.9	81.6	—	—	—	—	—	—	low
52—	M Kimi K2-Thinking-0905 Moonshot AI🇨🇳11moOSS	74.8	67.3	84.2	83.4	—	—	—	—	—	—	medium
53—	T Hy3 Tencent🇨🇳1moOSS	73.3	63.4	—	87.8	—	—	—	—	—	—	medium
54—	Midjourney v6.1 Midjourney🇺🇸2y	73.0	—	—	—	—	—	—	—	—	—	low
55—	M Kimi K2.6 Moonshot AI🇨🇳3moOSS	72.9	72.2	—	74.4	—	—	81.4	$4	—	—	high
56—	S Step-3.5-Flash StepFun🇨🇳6moOSS	72.9	70.3	80.9	—	—	—	—	$0.40	—	—	low
57—	GPT-5.2 OpenAI🇺🇸8mo	72.8	80.9	71.4	69.7	74.2	—	81.9	$14	—	—	high
58—	Z GLM-5 Zhipu AI🇨🇳6moOSS	72.8	77.0	—	—	—	—	—	$3	—	—	low
59—	DeepSeek-V4-Pro-Max DeepSeek🇨🇳3moOSS	72.7	58.5	—	84.9	78.1	—	—	$3	—	—	high
60—	o1-pro OpenAI🇺🇸1y	72.3	—	—	72.3	—	—	—	—	—	—	low
61—	Gemini 3.1 Pro Google🇺🇸5mo	72.3	68.8	—	89.3	52.3	—	81.7	$15	—	—	high
62—	Gemini 3 Pro Google🇺🇸8mo	72.2	74	84.2	65.7	64.5	—	75.3	—	—	—	high
63—	GPT-5.5 OpenAI🇺🇸3mo	72.1	53.6	51.0	90.9	83.6	—	86.3	$30	—	—	high
64—	GPT OSS 20B High OpenAI🇺🇸1yOSS	71.9	—	82.7	63.8	—	—	—	—	—	—	low
65—	Claude Opus 4.5 Anthropic🇺🇸8mo	71.8	82.3	—	61.3	78.0	—	—	—	—	—	medium
66—	Q Qwen3.7-Plus Alibaba Cloud / Qwen Team🇨🇳2mo	71.3	67.8	—	72.5	72.2	—	81.3	$1	—	—	high
67—	X MiMo-V2-Omni Xiaomi🇨🇳4mo	71.2	71.2	—	—	—	—	—	—	—	—	low
68—	M Kimi K2.5 Moonshot AI🇨🇳6moOSS	71.1	52.5	79.3	84.9	—	—	69.2	—	—	—	high
69—	S Step3-VL-10B StepFun🇨🇳7moOSS	70.6	—	66.3	—	—	—	77.2	—	—	—	low
70—	GPT-5.5 Instant OpenAI🇺🇸3mo	70.6	—	54.4	82.2	—	—	71.6	$30	—	—	medium
71—	GPT-5 Codex OpenAI🇺🇸11mo	70.5	70.5	—	—	—	—	—	—	—	—	low
72—	Claude Opus 4.1 Anthropic🇺🇸1y	70.3	75.0	48.4	75.4	73.8	—	—	—	—	—	medium
73—	OpenAI Codex OpenAI🇺🇸1y	69.7	58.4	—	—	—	—	—	—	—	—	low
74—	Q Qwen3.5-122B-A10B Alibaba Cloud / Qwen Team🇨🇳5moOSS	69.6	65.1	—	82.4	63.7	—	72.2	—	—	—	high
75—	Gemini 3 Flash Google🇺🇸8mo	69.4	77.4	83.9	65.1	62.2	—	71.7	$3	—	—	high
76—	MiniMax M3 MiniMax🇨🇳2moOSS	69.4	68.8	—	—	—	—	76.8	$1	—	—	medium
77—	GPT-5.1 Codex OpenAI🇺🇸8mo	68.9	68.9	—	—	—	—	—	—	—	—	low
78—	Q Qwen3.5-27B Alibaba Cloud / Qwen Team🇨🇳5moOSS	68.8	66.0	—	82.4	60.6	—	71.9	$2	—	—	high
79—	Claude Sonnet 4 Anthropic🇺🇸1y	68.7	54.1	66.8	71.2	90.0	—	63.7	$15	—	—	high
80—	U Solar Pro 4 Upstage🇰🇷1w	68.4	61.9	—	86.3	—	—	—	$1	—	—	low
81—	X Grok 4 Fast xAI🇺🇸11mo	68.2	—	73.4	54.4	98.7	—	—	—	—	—	medium
82—	GPT-5.4 OpenAI🇺🇸5mo	68.1	48.3	69.1	81.7	—	—	83.0	$15	—	—	high
83—	GPT OSS 120B High OpenAI🇺🇸1yOSS	68.0	—	74.1	75.4	52.1	—	—	—	—	—	low
84—	M Kimi K2.7 Code Moonshot AI🇨🇳2moOSS	67.6	—	—	—	—	—	—	$4	—	—	low
85—	MAI-Thinking-1 Microsoft🇺🇸2mo	67.6	45.2	80.5	80.3	—	—	—	—	—	—	medium
86—	Gemma 4 31B Google🇺🇸4moOSS	67.4	—	—	59.9	72.9	—	74.1	$0.38	—	—	medium
87—	DeepSeek-V3.2 DeepSeek🇨🇳8moOSS	67.4	67.6	75.0	73.8	—	—	—	—	—	—	medium
88—	L K-EXAONE-236B-A23B LG AI Research🇰🇷7mo	67.2	—	74.6	—	59.8	—	—	—	—	—	low
89—	Q Qwen3.5-397B-A17B Alibaba Cloud / Qwen Team🇨🇳5moOSS	67.1	74.4	—	64.9	70.4	—	—	—	—	—	medium
90—	Nova 2 Pro Amazon🇺🇸8mo	66.9	67.0	73.8	76.2	—	—	37.9	—	—	—	medium
91—	Z GLM-4.7 Zhipu AI🇨🇳7moOSS	66.9	56.8	78.8	78.0	—	—	—	—	—	—	medium
92—	Z GLM-5.1 Zhipu AI🇨🇳4moOSS	66.8	52.4	—	85.3	—	—	—	$4	—	—	medium
93—	M Kimi K2 0905 Moonshot AI🇨🇳11mo	66.7	—	—	66.7	—	—	—	—	—	—	low
94—	Q Qwen3-Next-80B-A3B-Thinking Alibaba Cloud / Qwen Team🇨🇳11moOSS	66.5	—	66.4	69.3	—	—	—	—	—	—	low
95—	MAI-Code-1.1-Flash Microsoft🇺🇸3d	66.5	66.5	—	—	—	—	—	$1	—	—	low
96—	M Muse Spark Meta🇺🇸4mo	66.5	48.3	—	75.5	—	—	84.8	—	—	—	high
97—	Llama 3.1 405B Meta🇺🇸2yOSS	66.3	69.3	72.0	48.6	67.9	87.7	—	$3	—	—	high
98—	Gemini 2.0 Flash Thinking Google🇺🇸1y	66.2	—	—	63.8	—	—	71.1	—	—	—	low
99—	Q Qwen3.6 Plus Alibaba Cloud / Qwen Team🇨🇳4mo	65.5	60.3	—	66.0	73.8	—	75.3	$3	—	—	high
100—	Q Qwen3.5-35B-A3B Alibaba Cloud / Qwen Team🇨🇳5moOSS	65.2	58.7	—	80.8	57.8	—	69.2	—	—	—	high