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.

441

Models Ranked

Benchmarks

Oct 1, 2026

Last Updated

All(441)LLMs(419)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)

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

Claude Opus 5.599.9 2

Claude Sonnet 5.599.0 3

Claude Fable 597.0 4

Claude Fable 5.196.6 5

Claude Opus 596.5

Research & Analysis

Best for scientific research, data analysis, and complex reasoning

Claude Fable 5.196.6 2

Content & Copywriting

Best for writing, editing, summarization, and creative tasks

Claude Fable 5.196.6 2

Best Value

Highest intelligence per dollar — the cost-efficiency sweet spot

DeepSeek-V4-Flash-0731495.1 2

Step-3.5-Flash181.9 3

GLM-5.3-Flash178.9 4

Qwen3.8 Flash160.3 5

Hy3125.8

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—	Claude Opus 5.5 Anthropic🇺🇸1w	99.9	99.9	—	—	—	—	—	$20	—	—	low
2—	X Grok 4.6 xAI🇺🇸1mo	99.6	—	—	—	—	—	—	$6	—	—	low
3—	Claude Sonnet 5.5 Anthropic🇺🇸3d	99.0	99.0	—	—	—	—	—	$10	—	—	low
4—	GPT-Image-1 OpenAI🇺🇸1y	98.7	—	—	—	—	—	—	—	—	—	low
5—	Claude Fable 5 Anthropic🇺🇸3mo	97.0	97.5	—	96.4	—	—	—	$50	—	—	low
6—	Claude Fable 5.1 Anthropic🇺🇸4w	96.6	—	—	96.6	—	—	—	$50	—	—	low
7—	M Muse Spark 1.3 Meta🇺🇸4w	96.0	—	—	—	96.0	—	—	$4	—	—	low
8—	Claude Opus 5 Anthropic🇺🇸2mo	95.2	—	—	96.5	—	—	—	$25	—	—	low
9—	Z GLM-5.3 Zhipu AI🇨🇳1moOSS	94.8	—	—	95.4	—	—	—	$4	—	—	low
10—	GPT-6 Astra OpenAI🇺🇸3w	94.2	—	—	94.1	—	—	96.3	$50	—	—	medium
11—	DeepSeek-V4-Pro-0813 DeepSeek🇨🇳1moOSS	93.8	—	—	93.9	—	—	—	$3	—	—	low
12—	Claude Mythos Preview Anthropic🇺🇸	91.4	96.7	—	93.8	83.8	—	90.2	—	—	—	medium
13—	Gemini 3.7 Flash Google🇺🇸1mo	90.7	—	—	—	95.5	—	83.5	$4	—	—	low
14—	Z GLM-5.3-Flash Zhipu AI🇨🇳1moOSS	89.5	—	—	90.1	—	—	84.7	$0.50	—	—	low
15—	Sora OpenAI🇺🇸1y	89.4	—	—	—	—	—	—	—	—	—	low
16—	DeepSeek-V4-Flash-0731 DeepSeek🇨🇳2moOSS	89.1	—	—	—	—	—	—	$0.18	—	—	low
17—	Claude Opus 4.8 Anthropic🇺🇸4mo	89.1	89.4	—	91.3	—	—	89.3	$25	—	—	high
18—	M Kimi K3 Moonshot AI🇨🇳2moOSS	88.2	—	—	90.4	—	—	85.8	$14	—	—	high
19—	Veo 2 Google🇺🇸1y	88.2	—	—	—	—	—	—	—	—	—	low
20—	GPT-5.6 Sol OpenAI🇺🇸2mo	87.4	74.4	98.2	91.0	93.3	—	86.3	$30	—	—	medium
21—	Q Qwen3.8 Max Alibaba Cloud / Qwen Team🇨🇳2moOSS	86.3	83.3	—	81.6	94.0	—	87.4	$5	—	—	medium
22—	GPT-5.6 Terra OpenAI🇺🇸2mo	84.9	70.4	97.4	89.6	92.4	—	82.3	$12	—	—	medium
23—	X Grok-4 Heavy xAI🇺🇸	84.7	—	84.3	84.9	—	—	—	—	—	—	low
24—	GPT-5.1 High OpenAI🇺🇸10mo	84.3	—	83.8	84.5	—	—	—	—	—	—	low
25—	Claude Opus 4.7 Anthropic🇺🇸5mo	84.2	82.3	—	90.1	80.6	—	87.2	$25	—	—	high
26—	Claude Sonnet 5 Anthropic🇺🇸3mo	83.5	79.1	—	92.0	—	—	82.8	$10	—	—	medium
27—	T Hy4 preview Tencent🇨🇳1moOSS	83.2	77.8	—	89.0	—	—	—	—	—	—	medium
28—	GPT-5.1 Medium OpenAI🇺🇸10mo	82.4	—	82.4	—	—	—	—	$10	—	—	low
29—	X Grok 4.5 xAI🇺🇸2mo	82.2	74.7	—	89.7	—	—	—	$6	—	—	low
30—	M Muse Spark 1.1 Meta🇺🇸2mo	81.3	63.6	—	95.2	—	—	83.0	$4	—	—	medium
31—	DALL-E 3 OpenAI🇺🇸3y	81.0	—	—	—	—	—	—	—	—	—	low
32—	GPT-5 High OpenAI🇺🇸1y	80.8	—	77.2	83.6	—	—	—	—	—	—	low
33—	B Seed 2.1 Pro ByteDance🇨🇳3mo	80.8	72.4	—	90.5	—	—	82.5	—	—	—	high
34—	Claude Opus 4.6 Anthropic🇺🇸7mo	80.2	82.4	84.1	85.2	81.1	—	66.0	$25	—	—	high
35—	GPT-5.1 Codex High OpenAI🇺🇸10mo	80.2	—	80.2	—	—	—	—	—	—	—	low
36—	B Seed 2.0 Pro ByteDance🇨🇳7mo	80.0	74.8	82.3	85.5	—	—	—	$3	—	—	medium
37—	ChatGPT-4o Latest OpenAI🇺🇸2y	79.1	—	—	79.1	—	—	—	—	—	—	low
38—	Gemini 3.8 Flash Google🇺🇸4w	78.8	—	—	—	—	—	78.8	$4	—	—	low
39—	Claude Sonnet 4.5 Anthropic🇺🇸1y	78.7	92.8	64.7	78.3	72.9	—	—	$15	—	—	medium
40—	GPT-5.5 Pro OpenAI🇺🇸5mo	78.6	—	57.5	91.8	—	—	—	—	—	—	low
41—	Claude Code Anthropic🇺🇸1y	78.6	66.8	—	—	—	—	—	—	—	—	low
42—	B Seed 2.1 Turbo ByteDance🇨🇳3mo	78.4	70.4	—	89.4	—	—	79.0	—	—	—	high
43—	X MiMo-V2-Pro Xiaomi🇨🇳6mo	77.6	77.6	—	—	—	—	—	—	—	—	low
44—	X Grok-3 xAI🇺🇸1y	77.5	—	75.3	80.0	—	—	75.9	$15	—	—	low
45—	GPT-5 Medium OpenAI🇺🇸1y	77.5	—	68.0	84.5	—	—	—	—	—	—	low
46—	Q Qwen3.7 Max Alibaba Cloud / Qwen Team🇨🇳4mo	76.9	76.5	—	79.4	76.9	—	—	$4	—	—	medium
47—	Z GLM-5V-Turbo Zhipu AI🇨🇳6mo	76.5	—	—	—	—	—	—	—	—	—	low
48—	ERNIE 5.0 Baidu🇨🇳8mo	76.3	—	64.7	72.8	92.5	—	—	—	—	—	medium
49—	GPT-5.2 Pro OpenAI🇺🇸9mo	76.2	—	84.3	70.8	—	—	—	—	—	—	medium
50—	X Grok-3 Mini xAI🇺🇸1y	75.8	—	71.3	79.1	—	—	—	—	—	—	low
51—	GPT-5.1 OpenAI🇺🇸10mo	75.8	74.4	57.1	84.5	—	—	89.0	$10	—	—	medium
52—	GPT-5.1 Instant OpenAI🇺🇸10mo	75.8	74.4	57.1	84.5	—	—	89.0	$10	—	—	medium
53—	GPT-5.1 Thinking OpenAI🇺🇸10mo	75.8	74.4	57.1	84.5	—	—	89.0	—	—	—	medium
54—	Claude Sonnet 4.6 Anthropic🇺🇸7mo	75.7	80.4	—	79.0	73.6	—	71.0	$15	—	—	high
55—	Q Qwen3.8 Flash Alibaba Cloud / Qwen Team🇨🇳1mo	75.3	67.2	—	73.4	—	—	86.6	$0.47	—	—	medium
56—	Q Qwen3.8-Flash-Next Alibaba Cloud / Qwen Team🇨🇳1moOSS	75.3	67.2	—	73.4	—	—	86.6	—	—	—	medium
57—	o1 OpenAI🇺🇸1y	75.2	48.4	94.2	79.3	86.7	—	—	$60	—	—	high
58—	M Kimi K2-Thinking-0905 Moonshot AI🇨🇳1yOSS	75.1	69.5	84.3	82.6	—	—	—	—	—	—	medium
59—	Claude Opus 4 Anthropic🇺🇸1y	74.8	58.0	75.5	79.3	96.0	—	66.8	$75	—	—	high
60—	B Seed 2.0 Lite ByteDance🇨🇳7mo	74.7	68.5	74.8	80.7	—	—	—	—	—	—	low
61—	Z GLM-5.2 Zhipu AI🇨🇳3moOSS	74.6	65.8	—	88.7	—	—	—	$2	—	—	medium
62—	DeepSeek-V4.1-Flash DeepSeek🇨🇳3wOSS	74.0	—	—	74.0	—	—	—	$0.66	—	—	low
63—	GPT-5.6 Luna OpenAI🇺🇸2mo	73.6	68.0	95.5	89.0	38.0	—	77.5	$1	—	—	medium
64—	DeepSeek-V4-Pro-Max DeepSeek🇨🇳5moOSS	73.1	60.9	—	84.1	78.1	—	—	$3	—	—	high
65—	Midjourney v6.1 Midjourney🇺🇸2y	73.0	—	—	—	—	—	—	—	—	—	low
66—	T Hy3 Tencent🇨🇳2moOSS	72.9	63.6	—	87.1	—	—	—	$0.58	—	—	medium
67—	S Step-3.5-Flash StepFun🇨🇳8moOSS	72.8	70.5	81.0	—	—	—	—	$0.40	—	—	low
68—	Z GLM-5 Zhipu AI🇨🇳7moOSS	72.7	77.2	—	—	—	—	—	$3	—	—	low
69—	Gemini 3.1 Pro Google🇺🇸7mo	72.4	71.2	—	87.9	51.8	—	81.9	$12	—	—	high
70—	GPT-5.2 OpenAI🇺🇸9mo	72.4	81.1	71.4	68.4	74.6	—	80.3	$14	—	—	high
71—	M Kimi K2.6 Moonshot AI🇨🇳5moOSS	72.3	72.7	—	73.3	—	—	80.4	$4	—	—	high
72—	L EXAONE 4.5 33B LG AI Research🇰🇷5mo	72.0	—	74.6	73.8	—	—	64.5	—	—	—	medium
73—	Gemini 3 Pro Google🇺🇸10mo	72.0	74.2	84.3	65.7	64.1	—	73.2	—	—	—	high
74—	Claude Opus 4.5 Anthropic🇺🇸10mo	71.9	82.6	—	61.3	78.5	—	—	—	—	—	medium
75—	M Kimi K2.5 Moonshot AI🇨🇳8moOSS	71.9	57.2	79.3	84.1	—	—	67.5	—	—	—	high
76—	X MiMo-V2-Omni Xiaomi🇨🇳6mo	71.3	71.3	—	—	—	—	—	—	—	—	low
77—	T Inkling Thinking Machines Lab🇺🇸2moOSS	71.3	76.9	—	—	—	—	62.2	$4	—	—	medium
78—	o1-pro OpenAI🇺🇸1y	71.2	—	—	71.2	—	—	—	—	—	—	low
79—	GPT OSS 20B High OpenAI🇺🇸1yOSS	71.2	—	82.8	62.5	—	—	—	—	—	—	low
80—	Q Qwen3.7-Plus Alibaba Cloud / Qwen Team🇨🇳4mo	71.2	69.0	—	71.4	72.6	—	79.6	—	—	—	high
81—	GPT-5.5 OpenAI🇺🇸5mo	70.9	52.3	51.0	89.7	80.8	—	86.6	$30	—	—	high
82—	GPT-5 Codex OpenAI🇺🇸1y	70.7	70.7	—	—	—	—	—	—	—	—	low
83—	Claude Opus 4.1 Anthropic🇺🇸1y	70	75.0	47.6	74.4	74.3	—	—	—	—	—	medium
84—	S Step3-VL-10B StepFun🇨🇳8moOSS	70.0	—	65.9	—	—	—	76.1	—	—	—	low
85—	OpenAI Codex OpenAI🇺🇸1y	69.7	58.4	—	—	—	—	—	—	—	—	low
86—	GPT-5.5 Instant OpenAI🇺🇸4mo	69.7	—	53.8	81.4	—	—	70.1	$30	—	—	medium
87—	Q Qwen3.8-27B Alibaba Cloud / Qwen Team🇨🇳1moOSS	69.5	64.3	—	66.5	—	—	86	$3	—	—	medium
88—	B Seed 1.8 ByteDance🇨🇳7mo	69.2	67.2	76.8	—	—	—	63.7	$2	—	—	medium
89—	Q Qwen3.5-122B-A10B Alibaba Cloud / Qwen Team🇨🇳7moOSS	69.0	65.2	—	81.5	64.0	—	70.3	$2	—	—	high
90—	Gemini 3 Flash Google🇺🇸9mo	69.0	77.6	84.0	65.0	62.0	—	69.5	$3	—	—	high
91—	GPT-5.1 Codex OpenAI🇺🇸10mo	69.0	69.0	—	—	—	—	—	—	—	—	low
92—	MAI-Thinking-1 Microsoft🇺🇸4mo	68.8	49.3	80.6	79.4	—	—	—	—	—	—	medium
93—	MiniMax M3 MiniMax🇨🇳4moOSS	68.7	67.9	—	—	—	—	76.9	$1	—	—	medium
94—	Claude Sonnet 4 Anthropic🇺🇸1y	68.3	53.9	66.8	70.5	90.0	—	62.3	$15	—	—	high
95—	Q Qwen3.5-27B Alibaba Cloud / Qwen Team🇨🇳7moOSS	68.2	66.1	—	81.6	60.8	—	70	$3	—	—	high
96—	U Solar Pro 4 Upstage🇰🇷1mo	67.9	62.0	—	85.6	—	—	—	$1	—	—	low
97—	M Muse Spark Meta🇺🇸5mo	67.7	52.6	—	75.0	—	—	83.3	—	—	—	high
98—	X Grok 4 Fast xAI🇺🇸1y	67.7	—	73.2	53.3	98.7	—	—	—	—	—	medium
99—	GPT OSS 120B High OpenAI🇺🇸1yOSS	67.6	—	74.0	74.4	52.2	—	—	—	—	—	low
100—	GPT-5.4 OpenAI🇺🇸7mo	67.6	48.7	69.1	80.0	—	—	83.2	$15	—	—	high