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.

414

Models Ranked

Benchmarks

Sep 5, 2026

Last Updated

All(414)LLMs(392)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

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 597.3 2

Claude Fable 5.196.6 3

Claude Opus 596.4 4

Muse Spark 1.395.7 5

GLM-5.395.4

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-0731496.8 2

Muse Spark 1.3478.6 3

Step-3.5-Flash182.5 4

GLM-5.3-Flash178.6 5

Gemma 4 31B174.7

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🇺🇸3w	99.4	—	—	—	—	—	—	$6	—	—	low
3—	Claude Fable 5 Anthropic🇺🇸2mo	97.2	98.0	—	96.3	—	—	—	$50	—	—	low
4—	Claude Fable 5.1 Anthropic🇺🇸4d	96.6	—	—	96.6	—	—	—	$50	—	—	low
5—	M Muse Spark 1.3 Meta🇺🇸3d	95.7	—	—	—	95.7	—	—	$0.20	—	—	low
6—	Claude Opus 5 Anthropic🇺🇸1mo	95.1	—	—	96.4	—	—	—	$25	—	—	low
7—	Z GLM-5.3 Zhipu AI🇨🇳3wOSS	94.8	—	—	95.4	—	—	—	$4	—	—	low
8—	GPT-6 Astra OpenAI🇺🇸1d	94.2	—	—	94.2	—	—	96.3	$50	—	—	medium
9—	DeepSeek-V4-Pro-0813 DeepSeek🇨🇳3wOSS	93.8	—	—	93.9	—	—	—	$0.87	—	—	low
10—	Claude Mythos Preview Anthropic🇺🇸	91.4	97.4	—	93.8	83.3	—	89.5	—	—	—	medium
11—	Gemini 3.7 Flash Google🇺🇸3w	90.2	—	—	—	95.2	—	82.8	$4	—	—	low
12—	Claude Opus 4.8 Anthropic🇺🇸3mo	89.8	91.7	—	91.3	—	—	88.9	$25	—	—	high
13—	DeepSeek-V4-Flash-0731 DeepSeek🇨🇳1moOSS	89.4	—	—	—	—	—	—	$0.18	—	—	low
15—	Z GLM-5.3-Flash Zhipu AI🇨🇳1wOSS	89.3	—	—	90.0	—	—	84.0	$0.50	—	—	low
16—	GPT-5.6 Sol OpenAI🇺🇸1mo	88.7	80.0	98.2	91.2	93.0	—	86.2	$30	—	—	medium
18—	M Kimi K3 Moonshot AI🇨🇳1moOSS	88.0	—	—	90.5	—	—	85.4	$15	—	—	high
19—	Q Qwen3.8 Max Alibaba Cloud / Qwen Team🇨🇳1moOSS	87.7	88.3	—	81.7	93.6	—	87.3	$5	—	—	medium
20—	GPT-5.6 Terra OpenAI🇺🇸1mo	86.1	76.1	97.4	89.8	92.0	—	82.2	$12	—	—	medium
21—	T Hy4 preview Tencent🇨🇳1wOSS	85.6	83.3	—	89.2	—	—	—	—	—	—	medium
22—	X Grok 4.5 xAI🇺🇸1mo	85.1	80.3	—	89.8	—	—	—	$6	—	—	low
23—	Claude Opus 4.7 Anthropic🇺🇸4mo	84.8	85.1	—	90.2	80.1	—	86.5	$25	—	—	high
24—	X Grok-4 Heavy xAI🇺🇸	84.7	—	84.2	85.0	—	—	—	—	—	—	low
25—	Claude Sonnet 5 Anthropic🇺🇸2mo	84.3	82.0	—	91.9	—	—	82.0	$10	—	—	medium
26—	GPT-5.1 High OpenAI🇺🇸9mo	84.3	—	83.8	84.8	—	—	—	—	—	—	low
27—	M Muse Spark 1.1 Meta🇺🇸1mo	83.2	69.0	—	95.2	—	—	82.2	$4	—	—	medium
28—	GPT-5.1 Medium OpenAI🇺🇸9mo	82.3	—	82.3	—	—	—	—	$10	—	—	low
29—	B Seed 2.1 Pro ByteDance🇨🇳2mo	81.3	74.5	—	90.4	—	—	82.0	—	—	—	high
30—	GPT-5 High OpenAI🇺🇸1y	81.0	—	77.2	83.8	—	—	—	—	—	—	low
32—	B Seed 2.0 Pro ByteDance🇨🇳6mo	80.2	75.1	82.2	85.7	—	—	—	$3	—	—	medium
33—	GPT-5.1 Codex High OpenAI🇺🇸9mo	80.1	—	80.1	—	—	—	—	—	—	—	low
34—	Claude Opus 4.6 Anthropic🇺🇸7mo	80.0	82.6	84	84.8	80.6	—	65.9	$25	—	—	high
35—	ChatGPT-4o Latest OpenAI🇺🇸2y	79.4	—	—	79.4	—	—	—	—	—	—	low
36—	B Seed 2.1 Turbo ByteDance🇨🇳2mo	78.9	72.4	—	89.4	—	—	78.6	—	—	—	high
38—	Claude Sonnet 4.5 Anthropic🇺🇸11mo	78.7	92.8	64.8	78.5	72.5	—	—	$15	—	—	medium
39—	GPT-5.5 Pro OpenAI🇺🇸4mo	78.6	—	57.5	91.8	—	—	—	—	—	—	low
40—	Gemini 3.8 Flash Google🇺🇸3d	78.0	—	—	—	—	—	78.0	$4	—	—	low
41—	X MiMo-V2-Pro Xiaomi🇨🇳5mo	77.9	77.9	—	—	—	—	—	—	—	—	low
42—	X Grok-3 xAI🇺🇸1y	77.8	—	75.2	80.3	—	—	76.7	$15	—	—	low
43—	GPT-5 Medium OpenAI🇺🇸1y	77.6	—	68.1	84.8	—	—	—	—	—	—	low
44—	Q Qwen3.7 Max Alibaba Cloud / Qwen Team🇨🇳3mo	77.4	78.3	—	79.5	76.5	—	—	$4	—	—	medium
45—	Z GLM-5V-Turbo Zhipu AI🇨🇳5mo	77.3	—	—	—	—	—	—	—	—	—	low
46—	Q Qwen3.8 Flash Alibaba Cloud / Qwen Team🇨🇳1w	77.3	72.9	—	73.5	—	—	85.9	$0.47	—	—	medium
47—	Q Qwen3.8-Flash-Next Alibaba Cloud / Qwen Team🇨🇳1wOSS	77.3	72.9	—	73.5	—	—	85.9	—	—	—	medium
48—	Z GLM-5.2 Zhipu AI🇨🇳2moOSS	76.9	71.4	—	88.8	—	—	—	$3	—	—	medium
49—	ERNIE 5.0 Baidu🇨🇳7mo	76.3	—	64.8	72.9	92.5	—	—	—	—	—	medium
50—	GPT-5.1 OpenAI🇺🇸9mo	76.0	74.7	57.0	84.8	—	—	89.5	$10	—	—	medium
51—	GPT-5.1 Instant OpenAI🇺🇸9mo	76.0	74.7	57.0	84.8	—	—	89.5	$10	—	—	medium
52—	GPT-5.1 Thinking OpenAI🇺🇸9mo	76.0	74.7	57.0	84.8	—	—	89.5	—	—	—	medium
53—	X Grok-3 Mini xAI🇺🇸1y	75.9	—	71.3	79.4	—	—	—	—	—	—	low
54—	GPT-5.2 Pro OpenAI🇺🇸8mo	75.8	—	84.2	69.9	—	—	—	—	—	—	medium
55—	Claude Sonnet 4.6 Anthropic🇺🇸6mo	75.5	80.7	—	78.2	73.2	—	71.1	$15	—	—	high
56—	o1 OpenAI🇺🇸1y	75.3	48.6	94.2	79.5	86.7	—	—	$60	—	—	high
57—	M Kimi K2-Thinking-0905 Moonshot AI🇨🇳12moOSS	75.3	69.6	84.2	82.7	—	—	—	—	—	—	medium
58—	Claude Opus 4 Anthropic🇺🇸1y	75.1	58.3	75.5	79.5	96.0	—	67.8	$75	—	—	high
59—	GPT-5.6 Luna OpenAI🇺🇸1mo	74.9	73.6	95.5	89.2	38.4	—	77.6	$1	—	—	medium
60—	B Seed 2.0 Lite ByteDance🇨🇳6mo	74.9	68.9	74.8	81.0	—	—	—	—	—	—	low
61—	T Hy3 Tencent🇨🇳2moOSS	73.9	65.8	—	87.3	—	—	—	—	—	—	medium
62—	DeepSeek-V4-Pro-Max DeepSeek🇨🇳4moOSS	73.6	62.5	—	84.2	78.1	—	—	—	—	—	high
63—	S Step-3.5-Flash StepFun🇨🇳7moOSS	73.0	70.8	80.9	—	—	—	—	$0.40	—	—	low
65—	Z GLM-5 Zhipu AI🇨🇳6moOSS	73	77.5	—	—	—	—	—	$3	—	—	low
66—	M Kimi K2.6 Moonshot AI🇨🇳4moOSS	72.8	74.3	—	73.5	—	—	80.0	$4	—	—	high
67—	Gemini 3.1 Pro Google🇺🇸6mo	72.6	72.0	—	87.7	51.8	—	81.8	$15	—	—	high
68—	L EXAONE 4.5 33B LG AI Research🇰🇷4mo	72.3	—	74.6	74.1	—	—	65.2	—	—	—	medium
69—	GPT-5.2 OpenAI🇺🇸8mo	72.1	81.3	71.4	67.5	74.2	—	80.1	$14	—	—	high
70—	M Kimi K2.5 Moonshot AI🇨🇳7moOSS	72.0	57.5	79.3	84.2	—	—	67.3	—	—	—	high
71—	GPT-5.5 OpenAI🇺🇸4mo	71.8	56.9	51.0	89.6	80.4	—	86.5	$30	—	—	high
72—	Gemini 3 Pro Google🇺🇸9mo	71.7	74.5	84.2	64.5	64.1	—	73	—	—	—	high
73—	X MiMo-V2-Omni Xiaomi🇨🇳5mo	71.7	71.7	—	—	—	—	—	—	—	—	low
74—	Q Qwen3.8-27B Alibaba Cloud / Qwen Team🇨🇳3wOSS	71.7	69.8	—	66.7	—	—	85.3	—	—	—	medium
75—	o1-pro OpenAI🇺🇸1y	71.6	—	—	71.6	—	—	—	—	—	—	low
76—	Q Qwen3.7-Plus Alibaba Cloud / Qwen Team🇨🇳3mo	71.5	70.5	—	71.5	72.2	—	79.3	—	—	—	high
77—	GPT OSS 20B High OpenAI🇺🇸1yOSS	71.4	—	82.7	62.9	—	—	—	—	—	—	low
78—	Claude Opus 4.5 Anthropic🇺🇸9mo	71.3	82.8	—	59.5	78.0	—	—	—	—	—	medium
79—	GPT-5 Codex OpenAI🇺🇸11mo	71.0	71.0	—	—	—	—	—	—	—	—	low
80—	S Step3-VL-10B StepFun🇨🇳7moOSS	70.4	—	66	—	—	—	76.9	—	—	—	low
81—	Claude Opus 4.1 Anthropic🇺🇸1y	70.1	75.2	47.8	74.7	73.8	—	—	—	—	—	medium
82—	MiniMax M3 MiniMax🇨🇳3moOSS	70.1	70.4	—	—	—	—	76.9	$1	—	—	medium
84—	GPT-5.5 Instant OpenAI🇺🇸4mo	69.8	—	54.0	81.6	—	—	69.8	$30	—	—	medium
85—	GPT-5.1 Codex OpenAI🇺🇸9mo	69.3	69.3	—	—	—	—	—	—	—	—	low
86—	MAI-Thinking-1 Microsoft🇺🇸3mo	69.3	50.4	80.5	79.7	—	—	—	—	—	—	medium
87—	Q Qwen3.5-122B-A10B Alibaba Cloud / Qwen Team🇨🇳6moOSS	69.2	65.6	—	81.6	63.7	—	70.4	—	—	—	high
88—	Gemini 3 Flash Google🇺🇸8mo	68.7	77.9	83.9	63.7	62.0	—	69.3	$3	—	—	high
89—	GPT-5.4 OpenAI🇺🇸6mo	68.7	52.9	69.1	79.7	—	—	83.1	$15	—	—	high
90—	Claude Sonnet 4 Anthropic🇺🇸1y	68.5	54.2	66.8	70.7	90.0	—	63.3	$15	—	—	high
91—	Q Qwen3.5-27B Alibaba Cloud / Qwen Team🇨🇳6moOSS	68.4	66.5	—	81.7	60.6	—	70.1	$2	—	—	high
92—	U Solar Pro 4 Upstage🇰🇷4w	68.3	62.4	—	85.8	—	—	—	$1	—	—	low
93—	Z GLM-5.1 Zhipu AI🇨🇳5moOSS	67.9	56.0	—	84.6	—	—	—	$4	—	—	medium
94—	X Grok 4 Fast xAI🇺🇸1y	67.8	—	73.2	53.5	98.7	—	—	—	—	—	medium
95—	GPT OSS 120B High OpenAI🇺🇸1yOSS	67.7	—	74	74.7	52.1	—	—	—	—	—	low
96—	M Muse Spark Meta🇺🇸5mo	67.5	53.5	—	73.8	—	—	83.0	—	—	—	high
97—	L K-EXAONE-236B-A23B LG AI Research🇰🇷8mo	67.1	—	74.5	—	59.8	—	—	—	—	—	low
98—	DeepSeek-V3.2 DeepSeek🇨🇳9moOSS	67.1	68.0	74.9	72.8	—	—	—	—	—	—	medium
99—	MAI-Code-1.1-Flash Microsoft🇺🇸3w	66.9	66.9	—	—	—	—	—	$1	—	—	low
100—	Q Qwen3.5-397B-A17B Alibaba Cloud / Qwen Team🇨🇳6moOSS	66.8	74.9	—	63.9	70.4	—	—	—	—	—	medium
101—	Nova 2 Pro Amazon🇺🇸9mo	66.7	67.3	73.7	75.5	—	—	37.8	—	—	—	medium
102—	Z GLM-4.7 Zhipu AI🇨🇳8moOSS	66.6	57.0	78.7	77.1	—	—	—	—	—	—	medium
103—	Gemma 4 31B Google🇺🇸5moOSS	66.4	—	—	58.9	71.1	—	74.2	$0.38	—	—	medium
104—	M Kimi K2.7 Code Moonshot AI🇨🇳2moOSS	66.4	—	—	—	—	—	—	—	—	—	low
105—	Llama 3.1 405B Meta🇺🇸2yOSS	66.2	69.3	72.0	48.4	67.9	87.7	—	$3	—	—	high
106—	Q Qwen3-Next-80B-A3B-Thinking Alibaba Cloud / Qwen Team🇨🇳12moOSS	66.0	—	66.2	68.4	—	—	—	—	—	—	low
107—	M Kimi K2 0905 Moonshot AI🇨🇳12mo	65.9	—	—	65.9	—	—	—	—	—	—	low