Reasoning

Artificial Analysis public leaderboard

Intelligence Index

Composite Artificial Analysis index across mathematics, science, coding, and reasoning evaluations.

Published source

Metric

Index score

Results shown

633

Publisher

Artificial Analysis

Snapshot

Fetched 2026-09-08T22:57:37.545Z

633 of 633 models

Higher is better

Claude Fable 5.1 (max with fallback)

Claude Fable 5.1 (xhigh with fallback)

GPT-6 Astra (max)

GPT-6 Astra (xhigh)

Claude Fable 5.1 (high with fallback)

GPT-6 Astra (high)

Claude Opus 5 (max)

Claude Fable 5 (with fallback)

GPT-6 Astra (medium)

Claude Opus 5 (xhigh)

Claude Fable 5.1 (medium with fallback)

Claude Opus 5 (high)

Muse Spark 1.3 (max)

GPT-5.6 Sol (max)

Claude Fable 5.1 (low with fallback)

GPT-6 Astra (low)

GPT-6 Astra (Non-reasoning)

Muse Spark 1.3 (xhigh)

Claude Opus 5 (medium)

GLM-5.3 (max)

Grok 4.6 (high)

Grok 4.6 (xhigh)

GPT-5.6 Sol (xhigh)

Kimi K3 (max)

Grok 4.6 (medium)

GPT-5.6 Sol (high)

GPT-5.6 Terra (max)

Qwen3.8-Flash-Next

Claude Opus 4.8 (max)

GLM-5.3-Flash

Gemini 3.8 Flash (high)

Claude Opus 4.7 (max)

Qwen3.8 Max

Qwen3.8 2.4T A95B

Gemini 3.8 Flash (medium)

Muse Spark 1.2 (xhigh)

Claude Opus 5 (low)

Gemini 3.7 Flash (medium)

GPT-5.6 Sol (medium)

Gemini 3.7 Flash (high)

Grok 4.5 (high)

GPT-5.4 (xhigh)

GLM-5.2 (max)

GPT-5.5 (xhigh)

Claude Sonnet 5 (max)

GPT-5.6 Terra (xhigh)

GPT-5.6 Luna (max)

GPT-5.5 (high)

Gemini 3.7 Flash (low)

DeepSeek V4 Pro 0813 (max)

Grok 4.6 (low)

Agnes 2.5 Pro Beta

DeepSeek V4 Flash Vision (max)

GPT-5.6 Luna (xhigh)

DeepSeek V4 Flash 0731 (max)

GPT-5.6 Terra (high)

Kimi K3 (low)

Gemini 3.6 Flash

Muse Spark 1.1 (xhigh)

GPT-5.5 (medium)

K2 Horizon 375B A23B

Qwen3.8 27B (xhigh)

GPT-5.6 Sol (low)

Gemini 3.8 Flash (low)

Gemini 3.5 Flash (medium)

Motif 3

Gemini 3.5 Flash

GPT-5.6 Luna (high)

GPT-5.6 Terra (medium)

GPT-5.3 Codex (xhigh)

Motif 3 (Beta)

Claude Opus 4.6 (max)

Kimi K2.6

Muse Spark

Claude Opus 4.7 (Non-reasoning, high)

DeepSeek V4 Pro (max)

GPT-5.5 (low)

Qwen3.8 27B (medium)

Claude Sonnet 4.6 (max)

GPT-5.2 (xhigh)

Apodex 1.1

Gemini 3.1 Pro Preview

DeepSeek V4 Pro (high)

Qwen3.7 Max

MiniMax-M3

Qwen3.8 27B (low)

Claude Opus 4.5

Claude Sonnet 5 (Non-reasoning)

MiMo-V2-Pro

GPT-5.2 Codex (xhigh)

Claude Sonnet 5 (medium)

Qwen3.6 Max Preview

GPT-5.6 Sol (Non-reasoning)

Nex-N2-Pro

Solar Pro 4

Gemini 3 Pro Preview (high)

GLM-5

GPT-5.6 Terra (low)

GPT-5.4 (low)

GLM-5.1

JT-4.1 Flash 236B A21B

Grok Build 0.1 0616

Quasar 438B (max)

Qwen3.6 Plus

GPT-5.5 Instant (June 2026)

Agnes 2.5 Pro Alpha

GLM-5-Turbo

GPT-5.2 (medium)

MiMo-V2.5-Pro

Claude Opus 4.6 (Non-reasoning, high)

Gemini 3 Flash

Kimi K2.7 Code

Inkling Small

Qwen3.7 Plus

GPT-5.6 Luna (medium)

Hy3

Grok 4.20 0309 v2

Inkling

Grok 4.3 (high)

Grok 4.20 0309

MiMo-V2-Omni-0327

Ling 3.0 Flash

GPT-5 Codex (high)

DeepSeek V4 Flash (high)

Grok 4.3 (medium)

Claude Sonnet 5 (low)

Solar Open2 250B

GPT-5.1 (high)

Claude Sonnet 4.6 (Non-reasoning, high)

DeepSeek V4 Flash (max)

GPT-5.4 mini (xhigh)

Grok 4.3 (low)

GLM-5.1 (Non-reasoning)

MiMo-V2-Omni

Gemini 3.5 Flash (minimal)

GPT-5.1 Codex (high)

Claude Opus 4.5 (Non-reasoning)

Kimi K2.6 (Non-reasoning)

GLM 5V Turbo

Kimi K2.5

Nemotron 3 Ultra

Claude Sonnet 4.6 (Non-reasoning, low)

MiniMax-M2.7

GPT-5.5 (Non-reasoning)

GPT-5 (high)

Qwen3.5 27B

GPT-5 (medium)

Claude 4.1 Opus

MiniMax-M2.5

Hy3-preview

A.X-K2

GPT-5.5 Instant (May 2026)

Gemini 3.5 Flash-Lite

Grok 4

MiMo-V2-Flash (Feb 2026)

GLM-5.2 (Non-reasoning)

Qwen3.8 27B

Gemini 3 Pro Preview (low)

MiMo-V2.5

GPT-5.6 Terra (Non-reasoning)

Qwen3.6 35B A3B

GLM-4.7

Kimi K2 Thinking

Qwen3.6 27B

o3-pro

GPT-5.6 Luna (low)

G9v3-39A5B

GLM-5 (Non-reasoning)

KAT-Coder-Pro V2

DeepSeek V3.2

Qwen3.5 397B A17B (Non-reasoning)

Qwen3 Max Thinking

GPT-5.4 nano (xhigh)

Claude 4.5 Sonnet

MiniMax-M2.1

DeepSeek V4 Pro (Non-reasoning)

MiMo-V2-Flash

GPT-5 (low)

Claude 4 Opus

GPT-5 mini (medium)

Qwen3.5 Omni Plus

GPT-5.1 Codex mini (high)

Grok 4.1 Fast

GPT-5.4 nano

Qwen3.6 27B (Non-reasoning)

GPT-5.4 mini (medium)

K-

K-EXAONE 2.0

LongCat 2.0

Step 3.7 Flash

Kimi K2.5 (Non-reasoning)

Qwen3.5 27B (Non-reasoning)

Claude 4.5 Sonnet (Non-reasoning)

Qwen3.5 35B A3B

Qwen3.5 397B A17B

Claude 4 Sonnet

DeepSeek V4 Flash (Non-reasoning)

JT-35B-Flash

MiniMax-M2

KAT-Coder-Pro V1

Claude 4.1 Opus (Non-reasoning)

GLM-4.6

MiMo-V2.5-Pro (Non-reasoning)

GPT-5.4 (Non-reasoning)

Muse Glimmer (high)

Grok 4 Fast

Gemini 3 Flash (Non-reasoning)

Qwen3.5 122B A10B (Non-reasoning)

Claude 3.7 Sonnet

Claude 4.5 Haiku

GPT-5 mini (high)

GLM-4.7 (Non-reasoning)

Ring-2.6-1T

Hy3-preview (Non-reasoning)

Ling-2.6-1T

GPT-5.2 (Non-reasoning)

Step 3.5 Flash 2603

Doubao Seed Code

GPT-5.6 Luna (Non-reasoning)

Gemma 4 26B A4B

Gemini 2.5 Pro

o4-mini (high)

Step 3.5 Flash

Claude 4 Opus (Non-reasoning)

Claude 4 Sonnet (Non-reasoning)

DeepSeek V3.2 Exp

Qwen3 Max Thinking (Preview)

Qwen3.5 122B A10B

MiMo-V2-Flash (Non-reasoning)

DeepSeek V3.2 (Non-reasoning)

Gemini 3.1 Flash-Lite

Qwen3 Max

Gemini 2.5 Flash (Sep)

Gemma 4 31B

Claude 4.5 Haiku (Non-reasoning)

DeepSeek V3.1 Terminus

Kimi K2 0905

Claude 3.7 Sonnet (Non-reasoning)

Qwen3.6 35B A3B (Non-reasoning)

Qwen3.5 35B A3B (Non-reasoning)

Gemini 2.5 Pro (Mar)

GLM-4.6 (Non-reasoning)

Mistral Medium 3.5

GLM-4.7-Flash

Granite 4.2 30B

Grok 3 mini Reasoning (high)

Grok 4.20 0309 (Non-reasoning)

Grok 4.3 (Non-reasoning)

Gemini 2.5 Pro (May)

DeepSeek V3.2 Speciale

K-

K-EXAONE

MiniCPM5-2B

ERNIE 5.0 Thinking Preview

Grok 4.20 0309 v2 (Non-reasoning)

Gemma 4 12B

Nova 2.0 Pro Preview (medium)

Grok Code Fast 1

DeepSeek V3.1 Terminus (Non-reasoning)

Command A+

Gemma 4 31B (Non-reasoning)

DeepSeek V3.2 Exp (Non-reasoning)

Apriel-v1.5-15B-Thinker

DeepSeek V3.1 (Non-reasoning)

Qwen3.5 9B

Nova 2.0 Omni (medium)

Nemotron 3.5 Lightning

Nemotron 3 Super

DeepSeek V3.1

Qwen3 VL 235B A22B (Reasoning)

Apriel-v1.6-15B-Thinker

Nova 2.0 Lite (high)

GPT-5.1 (Non-reasoning)

Qwen3.5 9B (Non-reasoning)

EXAONE 4.5 33B

Gemma 4 26B A4B (Non-reasoning)

Qwen3.5 4B

DeepSeek R1 0528

Gemini 2.5 Flash

GPT-5 nano (high)

North Mini Code

Nova 2.0 Pro Preview (low)

GLM-4.5

Kimi K2

Qwen3 235B A22B 2507

GPT-4.1

Qwen3 Max (Preview)

Nova 2.0 Lite (medium)

GPT-5 nano (medium)

Qwen3.5 Omni Flash

o3-mini

o1-pro

Granite 4.2 8B

Gemini 2.5 Flash (Sep) (Non-reasoning)

gpt-oss-120b (high)

JT-MINI

Grok 3

Seed-OSS-36B-Instruct

Qwen3 235B 2507 (Non-reasoning)

Qwen3 Coder 480B

Qwen3 VL 32B (Reasoning)

Ling 3.0 Tiny

Magistral Medium 1.2

Sonar Reasoning Pro

Nova 2.0 Lite (low)

HyperNova 60B 2605 (high)

MiniMax M1 80k

GPT-5.4 nano (Non-reasoning)

Nemotron Cascade 2 30B A3B

Gemini 2.5 Flash (Apr)

Mercury 2

K2 Think V2

LongCat Flash Lite

Mistral Small 4

GPT-5 (minimal)

DeepSeek R1 (Jan)

o1-preview

HyperCLOVA X SEED Think (32B)

Grok 4.1 Fast (Non-reasoning)

GLM-4.6V

K-

K-EXAONE (Non-reasoning)

Qwen3 Next 80B A3B (Reasoning)

GPT-5.4 mini (Non-reasoning)

Nova 2.0 Omni (low)

GLM-4.5-Air

Grok 4 Fast (Non-reasoning)

Mi:dm K 2.5 Pro

o3-mini (high)

Ring-1T

Trinity Large Thinking

G9v3-3B

Qwen3.5 4B (Non-reasoning)

INTELLECT-3

GLM-4.7-Flash (Non-reasoning)

GPT-5 (ChatGPT)

Solar Open 100B

Grok 3 Reasoning Beta

Gemini 2.5 Flash-Lite (Sep)

Nemotron 3 Nano Omni 30B A3B

gpt-oss-120b (low)

GPT-4.1 mini

Qwen3 Coder Next

MiniMax M1 40k

Nova 2.0 Pro Preview (Non-reasoning)

gpt-oss-20b (low)

Qwen3 VL 235B A22B

GPT-5 mini (minimal)

Mistral Medium 3.1

K2-V2 (high)

Gemini 2.5 Flash (Non-reasoning)

Qwen3 30B A3B 2507

o1-mini

DeepSeek V3 0324

Mistral Large 3

Ling 2.6 Flash

Qwen3 Next 80B A3B

Tri-21B-think Preview

Qwen3 Coder 30B A3B

GPT-4.5 (Preview)

DiffusionGemma 26B A4B

Qwen3 235B

QwQ-32B

Qwen3 VL 30B A3B (Reasoning)

Gemini 2.0 Flash Thinking exp. (Jan)

Devstral 2

Gemma 4 12B (Non-reasoning)

Gemini 2.5 Flash-Lite (Sep) (Non-reasoning)

Llama 4 Maverick

Motif-2-12.7B

Ling-1T

Nova Premier

Solar Pro 2 (Preview)

Granite 4.2 3B

Magistral Medium 1

Mistral Medium 3

gpt-oss-20b (high)

K2-V2 (medium)

Llama Nemotron Super 49B v1.5

Devstral Medium

Mistral Small 4 (Non-reasoning)

Tri-21B-Think

GPT-4o (Mar)

Gemini 2.0 Flash

Claude 3.5 Haiku

Llama 3.3 Nemotron Super 49B

Gemma 4 E4B

Nemotron 3 Nano

Qwen3 4B 2507

MiniCPM5-1B

Sarvam 105B (high)

Gemini 2.0 Pro Experimental

Nova 2.0 Lite (Non-reasoning)

Devstral Small (May)

Claude 3 Opus

MiniCPM5-1B (Non-reasoning)

Sonar Reasoning

Gemini 2.5 Flash (Apr) (Non-reasoning)

Magistral Small 1.2

Gemini 2.5 Flash-Lite

DeepSeek V3 (Dec)

GPT-4o (Nov)

Nanbeige4.1-3B

LFM2.5-2.6B

Qwen3 VL 32B

DeepSeek R1 Distill Qwen 32B

Devstral Small 2

GLM-4.6V (Non-reasoning)

Qwen3 235B (Non-reasoning)

Magistral Small 1

Gemini 2.0 Flash (exp)

EXAONE 4.0 32B

Qwen3 VL 8B (Reasoning)

Nova 2.0 Omni (Non-reasoning)

DeepSeek R1 0528 Qwen3 8B

Qwen2.5 Max

Qwen3 VL 30B A3B

Hermes 4 70B

Gemini 1.5 Pro (Sep)

Solar Pro 2 (Preview) (Non-reasoning)

DeepSeek R1 Distill Llama 70B

Claude 3.5 Sonnet (Oct)

DeepSeek R1 Distill Qwen 14B

Falcon-H1R-7B

GPT-4.1 nano

Solar Pro 3

Ling-flash-2.0

Gemma 4 E2B

Qwen3 Omni 30B A3B (Reasoning)

GPT-4o (Aug)

Qwen2.5 72B

Sonar

Step3 VL 10B

Llama 3.3 70B

Qwen3 30B

Sonar Pro

Devstral Small

QwQ 32B-Preview

GLM-4.5V

Mistral Large 2 (Nov)

Llama Nemotron Ultra

Qwen3 30B A3B 2507 (Non-reasoning)

ERNIE 4.5 300B A47B

Hermes 4 405B

Solar Pro 2

NVIDIA Nemotron Nano 12B v2 VL

Gemma 4 E4B (Non-reasoning)

Granite 4.1 30B

NVIDIA Nemotron Nano 9B V2

Hermes 4 405B (Non-reasoning)

Gemini 2.0 Flash-Lite (Feb)

Nemotron 3 Nano 4B

Mistral Small 3.1

Llama Nemotron Super 49B v1.5 (Non-reasoning)

Qwen3 32B (Non-reasoning)

GPT-4o (May)

Gemini 2.0 Flash-Lite (Preview)

K2-V2 (low)

Llama 3.1 Nemotron Nano 4B v1.1

Kimi Linear 48B A3B Instruct

Llama 3.1 405B

Llama 3.3 Nemotron Super 49B (Non-reasoning)

Qwen3 VL 8B

Qwen3 4B

LFM2.5-8B-A1B

Qwen3 32B

Claude 3.5 Sonnet (June)

Tulu3 405B

GPT-4o (ChatGPT)

Ring-flash-2.0

Pixtral Large

Olmo 3.1 32B Think

Grok 2

GPT-5 nano (minimal)

Gemini 1.5 Flash (Sep)

Qwen3 VL 4B (Reasoning)

GPT-4 Turbo

Solar Pro 2 (Non-reasoning)

Mistral Small 3.2

Nova Pro

Command A

Qwen3.5 2B

Llama 3.1 Nemotron 70B

Llama 3.1 8B

Grok Beta

Qwen2.5 Instruct 32B

Nemotron 3 Nano (Non-reasoning)

NVIDIA Nemotron Nano 9B V2 (Non-reasoning)

Mistral Large 2 (Jul)

Qwen3 4B 2507 (Non-reasoning)

Qwen2.5 Coder 32B

Qwen3 14B (Non-reasoning)

GPT-4

GLM-4.5V (Non-reasoning)

Mistral Small 3

Gemini 2.5 Flash-Lite (Non-reasoning)

Nova Lite

GPT-4o mini

Hermes 4 70B (Non-reasoning)

Qwen3 30B (Non-reasoning)

DeepSeek-V2.5 (Dec)

Qwen3 4B (Non-reasoning)

Llama 3.1 70B

Granite 4.1 8B

Sarvam 30B (high)

Gemini 2.0 Flash Thinking exp. (Dec)

DeepSeek-V2.5

Olmo 3.1 32B Instruct

Mistral Saba

DeepSeek R1 Distill Llama 8B

Gemma 4 E2B (Non-reasoning)

Olmo 3 32B Think

Llama 4 Scout

Gemini 1.5 Pro (May)

R1 1776

Qwen2.5 Turbo

Reka Flash

Llama 3.2 90B (Vision)

Solar Mini

Qwen3 14B

Celeris-1

Grok-1

Qwen2 72B

Phi-4 Mini

EXAONE 4.0 32B (Non-reasoning)

Qwen3.5 2B (Non-reasoning)

Gemini 1.5 Flash-8B

Qwen3.5 0.8B

DeepHermes 3 - Mistral 24B

Jamba 1.7 Large

Granite 4.0 H Small

Ministral 3 14B

Jamba 1.5 Large

Qwen3 Omni 30B A3B

Hermes 3 - Llama-3.1 70B

Qwen3 8B (Non-reasoning)

DeepSeek-Coder-V2

OLMo 2 32B

Jamba 1.6 Large

LFM2 24B A2B

Gemini 1.5 Flash (May)

Phi-4

Claude 3 Sonnet

Nova Micro

Granite 4.1 3B

Mistral Small (Sep)

Gemini 1.0 Ultra

Phi-3 Mini

NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)

Gemma 3n E4B (May)

Phi-4 Multimodal

Qwen2.5 Coder 7B

Mistral Large (Feb)

Mixtral 8x22B

Llama 2 Chat 7B

Llama 3.2 3B

MiniCPM-V 4.6 1.3B

Jamba Reasoning 3B

Qwen3 VL 4B

Qwen1.5 Chat 110B

Reka Flash 3

Olmo 3 7B Think

Claude 2.1

Claude 3 Haiku

OLMo 2 7B

Molmo 7B-D

Ling-mini-2.0

DeepSeek R1 Distill Qwen 1.5B

DeepSeek-V2

Claude 2.0

Mistral Small (Feb)

Mistral Medium

GPT-3.5 Turbo

Ministral 3 8B

Llama 3 70B

Qwen Chat 72B

Arctic

LFM 40B

Llama 3.2 11B (Vision)

Qwen3.5 0.8B (Non-reasoning)

PALM-2

Gemini 1.0 Pro

DeepSeek Coder V2 Lite

Sarvam M

Llama 2 Chat 70B

DeepSeek LLM 67B (V1)

Llama 2 Chat 13B

Command-R+ (Apr)

OpenChat 3.5

DBRX

Exaone 4.0 1.2B

Olmo 3 7B

Exaone 4.0 1.2B (Non-reasoning)

LFM2.5-1.2B-Thinking

Jamba 1.7 Mini

LFM2 2.6B

LFM2.5-1.2B-Instruct

Jamba 1.5 Mini

Granite 4.0 H 1B

Qwen3 1.7B

Jamba 1.6 Mini

Qwen3 8B

Mixtral 8x7B

Gemma 3 270M

Apertus 70B Instruct

Granite 4.0 Micro

DeepHermes 3 - Llama-3.1 8B

Mistral 7B

Claude Instant

Command-R (Mar)

Qwen Chat 14B

Llama 65B

Granite 4.0 1B

Molmo2-8B

LFM2 8B A1B

Granite 3.3 8B

Qwen3 1.7B (Non-reasoning)

Gemma 3 27B

Ministral 3 3B

Qwen3 0.6B (Non-reasoning)

LFM2 1.2B

Gemma 3 4B

LFM2.5-VL-1.6B

Qwen3 0.6B

Granite 4.0 350M

Llama 3.2 1B

Tiny Aya Global

Llama 3 8B

Apertus 8B Instruct

Granite 4.0 H 350M

Gemma 3n E4B

Gemma 3n E2B

Gemma 3 1B

Gemma 3 12B

List bars visualize relative performance. Plot bars use published 0–100 scores, shrink to a 1.5rem minimum, then scroll horizontally.

Scores are published snapshots and are only comparable within the same benchmark, version, metric, and evaluation configuration.