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.

459

Models Ranked

Benchmarks

Oct 21, 2026

Last Updated

All(459)LLMs(437)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)

Mistral AI

Anthropic

xAI

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

Mistral Large 4100.0 2

Claude Opus 5.599.9 3

Claude Sonnet 5.599.0 4

Claude Fable 597.1 5

Claude Fable 5.196.8

Research & Analysis

Best for scientific research, data analysis, and complex reasoning

Claude Fable 5.196.8 2

Content & Copywriting

Best for writing, editing, summarization, and creative tasks

Claude Fable 5.196.8 2

Best Value

Highest intelligence per dollar — the cost-efficiency sweet spot

DeepSeek-V4-Flash-0731495.1 2

Step-3.5-Flash181.3 3

GLM-5.3-Flash179.3 4

Qwen3.8 Flash160.4 5

Claude Haiku 5.5150.4

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—	Mistral Large 4 Mistral AI🇫🇷2w	100.0	100.0	—	—	—	—	—	$2	—	—	low
2—	Claude Opus 5.5 Anthropic🇺🇸4w	99.9	99.9	—	—	—	—	—	$20	—	—	low
3—	Claude Sonnet 5.5 Anthropic🇺🇸3w	99.0	99.0	—	—	—	—	—	$10	—	—	low
4—	GPT-Image-1 OpenAI🇺🇸1y	98.7	—	—	—	—	—	—	—	—	—	low
5—	X Grok 4.6 xAI🇺🇸2mo	97.1	—	—	—	—	—	—	$6	—	—	low
6—	Claude Fable 5 Anthropic🇺🇸4mo	97.0	97.5	—	96.6	—	—	—	$50	—	—	low
7—	Claude Fable 5.1 Anthropic🇺🇸1mo	96.8	—	—	96.8	—	—	—	$50	—	—	low
8—	M Muse Spark 1.3 Meta🇺🇸1mo	96.0	—	—	—	96.0	—	—	$4	—	—	low
9—	Claude Opus 5 Anthropic🇺🇸2mo	95.3	—	—	96.7	—	—	—	$25	—	—	low
10—	Z GLM-5.3 Zhipu AI🇨🇳2moOSS	95.0	—	—	95.7	—	—	—	$4	—	—	low
11—	F Ember-1 Fireworks AI🇺🇸4w	94.8	94.8	—	—	—	—	—	$15	—	—	low
12—	GPT-6 Astra OpenAI🇺🇸1mo	94.2	—	—	94.0	—	—	96.3	$50	—	—	medium
13—	DeepSeek-V4-Pro-0813 DeepSeek🇨🇳2moOSS	94.0	—	—	94.2	—	—	—	$3	—	—	low
14—	GPT-6.1 Sol OpenAI🇺🇸3w	92.8	—	—	—	—	—	90.1	$10	—	—	low
15—	Claude Mythos Preview Anthropic🇺🇸	91.4	96.7	—	93.8	83.8	—	90.2	—	—	—	medium
16—	Gemini 3.7 Flash Google🇺🇸2mo	90.7	—	—	—	95.5	—	83.4	$4	—	—	low
17—	Z GLM-5.3-Flash Zhipu AI🇨🇳1moOSS	89.7	—	—	90.5	—	—	84.6	$0.50	—	—	low
18—	Sora OpenAI🇺🇸1y	89.4	—	—	—	—	—	—	—	—	—	low
19—	DeepSeek-V4-Flash-0731 DeepSeek🇨🇳2moOSS	89.1	—	—	—	—	—	—	$0.18	—	—	low
20—	Claude Opus 4.8 Anthropic🇺🇸4mo	89.1	89.4	—	91.4	—	—	89.2	$25	—	—	high
21—	Veo 2 Google🇺🇸1y	88.2	—	—	—	—	—	—	—	—	—	low
22—	M Kimi K3 Moonshot AI🇨🇳3moOSS	87.7	—	—	90.5	—	—	85.5	$14	—	—	high
23—	GPT-5.6 Sol OpenAI🇺🇸3mo	87.4	74.6	98.2	90.9	93.3	—	85.8	$30	—	—	medium
24—	M Muse Spark Open Meta🇺🇸6dOSS	86.5	86	—	85.3	—	—	—	$4	—	—	low
25—	Q Qwen3.8 Max Alibaba Cloud / Qwen Team🇨🇳2moOSS	86.3	83.5	—	81.7	94.0	—	87.1	$5	—	—	medium
26—	GPT-5.6 Terra OpenAI🇺🇸3mo	84.8	70.6	97.4	89.4	92.4	—	81.7	$12	—	—	medium
27—	X Grok-4 Heavy xAI🇺🇸	84.7	—	84.3	85.0	—	—	—	—	—	—	low
28—	Claude Opus 4.7 Anthropic🇺🇸6mo	84.2	82.3	—	90.2	80.6	—	87.2	$25	—	—	high
29—	GPT-5.1 High OpenAI🇺🇸11mo	84.1	—	83.8	84.3	—	—	—	—	—	—	low
30—	Claude Sonnet 5 Anthropic🇺🇸3mo	83.6	79.1	—	92.3	—	—	82.7	$10	—	—	medium
31—	T Hy4 preview Tencent🇨🇳1moOSS	82.7	78.0	—	88.8	—	—	—	—	—	—	medium
32—	GPT-5.1 Medium OpenAI🇺🇸11mo	82.4	—	82.4	—	—	—	—	$10	—	—	low
33—	X Grok 4.5 xAI🇺🇸3mo	82.2	74.9	—	89.5	—	—	—	$6	—	—	low
34—	M Muse Spark 1.1 Meta🇺🇸3mo	81.4	63.7	—	95.5	—	—	82.9	$4	—	—	medium
35—	DALL-E 3 OpenAI🇺🇸3y	81.0	—	—	—	—	—	—	—	—	—	low
36—	GPT-5 High OpenAI🇺🇸1y	80.7	—	77.2	83.3	—	—	—	—	—	—	low
37—	GPT-5.1 Codex High OpenAI🇺🇸11mo	80.2	—	80.2	—	—	—	—	—	—	—	low
38—	Claude Opus 4.6 Anthropic🇺🇸8mo	80.1	82.1	84.1	85.3	81.1	—	65.4	$25	—	—	high
39—	B Seed 2.0 Pro ByteDance🇨🇳8mo	79.8	74.4	82.3	85.2	—	—	—	$3	—	—	medium
40—	B Seed 2.1 Pro ByteDance🇨🇳3mo	79.1	68.1	—	90.9	—	—	82.1	—	—	—	high
41—	ChatGPT-4o Latest OpenAI🇺🇸2y	78.8	—	—	78.8	—	—	—	—	—	—	low
42—	GPT-5.5 Pro OpenAI🇺🇸6mo	78.8	—	57.5	92.2	—	—	—	—	—	—	low
43—	Claude Sonnet 4.5 Anthropic🇺🇸1y	78.6	92.8	64.7	77.9	72.9	—	—	$15	—	—	medium
44—	Gemini 3.8 Flash Google🇺🇸1mo	78.6	—	—	—	—	—	78.6	$4	—	—	low
45—	Claude Code Anthropic🇺🇸1y	78.4	66.3	—	—	—	—	—	—	—	—	low
46—	X Grok-3 xAI🇺🇸1y	77.4	—	75.3	79.7	—	—	75.9	$15	—	—	low
47—	GPT-5 Medium OpenAI🇺🇸1y	77.3	—	68.0	84.3	—	—	—	—	—	—	low
48—	X MiMo-V2-Pro Xiaomi🇨🇳7mo	77.3	77.3	—	—	—	—	—	—	—	—	low
49—	Z GLM-5V-Turbo Zhipu AI🇨🇳6mo	76.5	—	—	—	—	—	—	—	—	—	low
50—	B Seed 2.1 Turbo ByteDance🇨🇳3mo	76.5	65.3	—	89.8	—	—	78.6	—	—	—	high
51—	ERNIE 5.0 Baidu🇨🇳9mo	76.3	—	64.7	72.8	92.5	—	—	—	—	—	medium
52—	GPT-5.2 Pro OpenAI🇺🇸10mo	76.2	—	84.3	70.8	—	—	—	—	—	—	medium
53—	X Grok-3 Mini xAI🇺🇸1y	75.6	—	71.3	78.8	—	—	—	—	—	—	low
54—	GPT-5.1 OpenAI🇺🇸11mo	75.5	74.0	57.1	84.3	—	—	89.0	$10	—	—	medium
55—	GPT-5.1 Instant OpenAI🇺🇸11mo	75.5	74.0	57.1	84.3	—	—	89.0	$10	—	—	medium
56—	GPT-5.1 Thinking OpenAI🇺🇸11mo	75.5	74.0	57.1	84.3	—	—	89.0	—	—	—	medium
57—	Claude Sonnet 4.6 Anthropic🇺🇸8mo	75.5	80.1	—	79.1	73.6	—	70.3	$15	—	—	high
58—	Q Qwen3.7 Max Alibaba Cloud / Qwen Team🇨🇳5mo	75.4	72.2	—	79.6	76.9	—	—	$4	—	—	medium
59—	Q Qwen3.8 Flash Alibaba Cloud / Qwen Team🇨🇳1mo	75.4	67.4	—	73.5	—	—	86.6	$0.47	—	—	medium
60—	Q Qwen3.8-Flash-Next Alibaba Cloud / Qwen Team🇨🇳1moOSS	75.4	67.4	—	73.5	—	—	86.6	—	—	—	medium
61—	Claude Haiku 5.5 Anthropic🇺🇸2w	75.2	75.2	—	—	—	—	—	$0.50	—	—	low
62—	o1 OpenAI🇺🇸1y	75.1	48.2	94.2	79.1	86.7	—	—	$60	—	—	high
63—	Claude Opus 4 Anthropic🇺🇸1y	74.7	57.8	75.5	79.0	96.0	—	66.8	$75	—	—	high
64—	Z GLM-5.2 Zhipu AI🇨🇳4moOSS	74.7	65.9	—	88.8	—	—	—	$2	—	—	medium
65—	B Seed 2.0 Lite ByteDance🇨🇳8mo	74.4	68.1	74.8	80.4	—	—	—	—	—	—	low
66—	DeepSeek-V4.1-Flash DeepSeek🇨🇳1moOSS	74.1	—	—	74.1	—	—	—	$0.66	—	—	low
67—	M Kimi K2-Thinking-0905 Moonshot AI🇨🇳1yOSS	74.1	66.5	84.3	82.6	—	—	—	—	—	—	medium
68—	GPT-5.6 Luna OpenAI🇺🇸3mo	73.5	68.1	95.5	88.8	38.0	—	76.9	$1	—	—	medium
69—	DeepSeek-V4-Pro-Max DeepSeek🇨🇳6moOSS	73.1	60.7	—	84.3	78.1	—	—	$3	—	—	high
70—	Midjourney v6.1 Midjourney🇺🇸2y	73.0	—	—	—	—	—	—	—	—	—	low
71—	T Hy3 Tencent🇨🇳3moOSS	72.7	63.4	—	86.9	—	—	—	$0.58	—	—	medium
72—	DeepSeek-V4.1-Flash-1017 DeepSeek🇨🇳4dOSS	72.5	—	—	72.5	—	—	—	$1	—	—	low
73—	S Step-3.5-Flash StepFun🇨🇳8moOSS	72.5	70.0	81.0	—	—	—	—	$0.40	—	—	low
74—	Z GLM-5 Zhipu AI🇨🇳8moOSS	72.5	76.9	—	—	—	—	—	$3	—	—	low
75—	GPT-5.2 OpenAI🇺🇸10mo	72.3	80.8	71.4	68.5	74.6	—	80.0	$14	—	—	high
76—	Gemini 3 Pro Google🇺🇸11mo	71.8	73.8	84.3	65.8	64.1	—	72.9	—	—	—	high
77—	Claude Opus 4.5 Anthropic🇺🇸11mo	71.8	82.3	—	61.1	78.5	—	—	—	—	—	medium
78—	L EXAONE 4.5 33B LG AI Research🇰🇷6mo	71.7	—	74.6	73.4	—	—	64.1	—	—	—	medium
79—	Gemini 3.1 Pro Google🇺🇸8mo	71.3	67.9	—	88.0	51.8	—	81.3	$12	—	—	high
80—	GPT OSS 20B High OpenAI🇺🇸1yOSS	70.9	—	82.8	62.1	—	—	—	—	—	—	low
81—	X MiMo-V2-Omni Xiaomi🇨🇳7mo	70.9	70.9	—	—	—	—	—	—	—	—	low
82—	GPT-5.5 OpenAI🇺🇸6mo	70.9	52.3	51.0	89.8	80.8	—	86.1	$30	—	—	high
83—	o1-pro OpenAI🇺🇸1y	70.8	—	—	70.8	—	—	—	—	—	—	low
84—	T Inkling Thinking Machines Lab🇺🇸3moOSS	70.8	76.5	—	—	—	—	61.6	$4	—	—	medium
85—	M Kimi K2.6 Moonshot AI🇨🇳6moOSS	70.7	68.3	—	73.5	—	—	80.0	$4	—	—	high
86—	M Kimi K2.5 Moonshot AI🇨🇳8moOSS	70.6	53.0	79.3	84.2	—	—	66.9	—	—	—	high
87—	GPT-5 Codex OpenAI🇺🇸1y	70.3	70.3	—	—	—	—	—	—	—	—	low
88—	MiniMax M3.1 Flash MiniMax🇨🇳2d	70.1	—	—	—	—	—	70.1	$0.60	—	—	low
89—	S Step3-VL-10B StepFun🇨🇳9moOSS	70.0	—	65.9	—	—	—	76.1	—	—	—	low
90—	Q Qwen3.7-Plus Alibaba Cloud / Qwen Team🇨🇳4mo	69.9	64.6	—	71.5	72.6	—	79.3	—	—	—	high
91—	Claude Opus 4.1 Anthropic🇺🇸1y	69.8	74.8	47.6	74.0	74.3	—	—	—	—	—	medium
92—	Q Qwen3.8-27B Alibaba Cloud / Qwen Team🇨🇳2moOSS	69.6	64.4	—	66.5	—	—	86.0	$3	—	—	medium
93—	OpenAI Codex OpenAI🇺🇸1y	69.5	57.9	—	—	—	—	—	—	—	—	low
94—	GPT-5.5 Instant OpenAI🇺🇸5mo	69.4	—	53.8	81.1	—	—	69.5	$30	—	—	medium
95—	Gemini 3 Flash Google🇺🇸10mo	69.0	77.3	84.0	65.0	62.0	—	69.2	$3	—	—	high
96—	Q Qwen3.5-122B-A10B Alibaba Cloud / Qwen Team🇨🇳7moOSS	68.9	64.7	—	81.6	64.0	—	70	$2	—	—	high
97—	B Seed 1.8 ByteDance🇨🇳8mo	68.9	66.8	76.8	—	—	—	63.2	$2	—	—	medium
98—	MAI-Thinking-1 Microsoft🇺🇸4mo	68.6	49.0	80.6	79.1	—	—	—	—	—	—	medium
99—	GPT-5.1 Codex OpenAI🇺🇸11mo	68.5	68.5	—	—	—	—	—	—	—	—	low
100—	MiniMax M3 MiniMax🇨🇳4moOSS	68.3	67.8	—	—	—	—	76.2	$1	—	—	medium