Instruction Following & Long Context

Artificial Analysis public leaderboard

Long Context Reasoning

Long-context reasoning results published by Artificial Analysis.

Published source

Metric

Accuracy

Results shown

508

Publisher

Artificial Analysis

Snapshot

Fetched 2026-08-22T12:55:16.136Z

508 of 508 models

Higher is better

Muse Spark 1.2 (xhigh)

Kimi K3 (max)

Muse Spark 1.1 (xhigh)

Gemini 3.5 Flash

Gemini 3.7 Flash (medium)

MiniMax-M3

Muse Glimmer (high)

Gemini 3.7 Flash (high)

Gemini 3.5 Flash (medium)

GPT-5.6 Terra (max)

GPT-5.2 Codex (xhigh)

GPT-5.2 (xhigh)

GPT-5.5 (xhigh)

Gemini 3.6 Flash

GPT-5.5 (high)

Gemini 3.1 Pro Preview

Grok 4.6 (low)

Claude Opus 5 (medium)

GPT-5.6 Luna (max)

GPT-5.3 Codex (xhigh)

Gemini 3.7 Flash (low)

MiMo-V2.5-Pro

GPT-5.4 (xhigh)

GPT-5.5 (low)

GPT-5.6 Sol (max)

Qwen3.8 27B (xhigh)

Claude Opus 5 (low)

Claude Sonnet 5 (max)

Kimi K3 (low)

Muse Spark

Kimi K2.6

GPT-5.1 (high)

GPT-5.5 (medium)

Claude Fable 5 (with fallback)

GLM-5.2 (max)

Claude Opus 5 (xhigh)

Nex-N2-Pro

GPT-5 (high)

GPT-5 (medium)

Qwen3.8 27B (medium)

Claude Opus 5 (high)

GLM-5.3 (max)

GPT-5.6 Sol (xhigh)

Claude Opus 4.5

Claude Opus 5 (max)

Grok 4.6 (xhigh)

MiniMax-M2.7

DeepSeek V4 Pro 0813 (max)

GPT-5.6 Sol (high)

Qwen3.8 2.4T A95B

Claude Opus 4.7 (max)

Kimi K2.7 Code

GPT-5.6 Terra (xhigh)

Grok 4.6 (high)

Gemini 3.5 Flash-Lite

Qwen3.7 Max

Qwen3.8 27B (low)

Hy3

Claude Opus 4.6 (max)

Qwen3.8 Max

GPT-5.6 Sol (medium)

DeepSeek V4 Flash 0731 (max)

Grok 4.5 (high)

GPT-5.6 Luna (high)

Claude Sonnet 4.6 (max)

GPT-5.4 (low)

Claude 4.5 Haiku

Inkling

GPT-5.6 Terra (high)

GPT-5.6 Luna (xhigh)

Qwen3.6 27B

Claude 4.1 Opus

GPT-5.6 Sol (low)

Gemini 3 Flash

Agnes 2.5 Pro Alpha

Claude Opus 4.8 (max)

Kimi K2.5

GPT-5.4 mini (xhigh)

Gemini 3 Pro Preview (high)

Qwen3.5 397B A17B

Grok 4.6 (medium)

Qwen3.5 27B

Claude Opus 4.7 (Non-reasoning, high)

DeepSeek V3.2 Exp

Motif 3

Qwen3.6 Plus

MiniMax-M2.5

GPT-5.6 Luna (medium)

Qwen3.6 Max Preview

GPT-5.4 nano (xhigh)

MiMo-V2-Omni

Gemini 3.1 Flash-Lite

GPT-5 mini (high)

Grok 4 Fast

Nemotron 3 Ultra

GPT-5 Codex (high)

Solar Pro 4

GLM-5

Motif 3 (Beta)

DeepSeek V3.2

Qwen3 235B A22B 2507

Qwen3.5 122B A10B

Grok 4.1 Fast

GPT-5.6 Terra (medium)

Qwen3 Max Thinking

JT-4.1 Flash 236B A21B

Gemini 3 Pro Preview (low)

Kimi K2 Thinking

Grok Build 0.1 0616

DeepSeek V4 Pro (max)

DeepSeek V4 Flash (max)

KAT-Coder-Pro V2

MiMo-V2-Omni-0327

Step 3.7 Flash

Claude 4 Sonnet

GPT-5 mini (medium)

Doubao Seed Code

Inkling Small

GPT-5.1 Codex (high)

GPT-5.2 (medium)

DeepSeek V4 Flash (high)

Qwen3.7 Plus

MiMo-V2-Flash (Feb 2026)

GPT-5.6 Terra (low)

MiMo-V2.5

Qwen3.5 35B A3B

Grok 4.3 (medium)

Solar Open2 250B

Claude 4.5 Sonnet

Gemma 4 31B

DeepSeek V3.1 Terminus

GLM-5.1

GLM-4.7

MiMo-V2-Flash

Gemini 2.5 Flash (Sep)

Grok 4.3 (low)

Claude Opus 4.5

Ring-2.6-1T

DeepSeek V4 Pro (high)

Grok 4

GPT-5.5 Instant (June 2026)

Ling 3.0 Flash

GLM-5-Turbo

Qwen3.6 35B A3B

MiniMax-M2.1

Kimi K2.6

Gemini 2.5 Pro

A.X-K2

Grok 4.3 (high)

Claude Sonnet 5 (Non-reasoning)

Gemini 2.5 Flash

GLM 5V Turbo

MiMo-V2-Pro

GPT-5.6 Luna (low)

Qwen3.5 9B

GPT-5.4 nano

Kimi K2.5

Mistral Medium 3.5

MiniMax-M2

GPT-5.1 Codex mini (high)

GPT-5 (ChatGPT)

Claude Sonnet 4.6 (Non-reasoning, Low Effort)

Qwen3.6 27B

GPT-4.1

DeepSeek V3.2 Speciale

GPT-5.5 Instant (May 2026)

JT-35B-Flash

GPT-5 (low)

Nova 2.0 Pro Preview (low)

Gemini 2.5 Flash-Lite (Sep)

Qwen3 VL 235B A22B

LongCat 2.0

Claude 3.7 Sonnet

Claude Sonnet 4.6 (Non-reasoning)

GPT-5.4 mini (medium)

Claude Opus 4.6 (high)

Grok 4.20 0309 v2

Qwen3.5 397B A17B

G9v3-39A5B

Qwen3 Next 80B A3B

Qwen3.5 122B A10B

Gemma 4 26B A4B

Gemma 4 12B

Qwen3.5 27B

Qwen3.5 4B

Grok 4.20 0309

Step 3.5 Flash 2603

Nemotron 3 Super

Hy3-preview

Qwen3 Max Thinking (Preview)

Qwen3.5 35B A3B

Nova 2.0 Lite (medium)

Seed-OSS-36B-Instruct

Qwen3.6 35B A3B

o4-mini (high)

Qwen3 30B A3B 2507

K-

K-EXAONE

Gemini 2.5 Flash (Sep)

Nova 2.0 Lite (high)

Ling 3.0 Tiny

Gemini 3.5 Flash (minimal)

Qwen3.5 Omni Plus

Nova 2.0 Pro Preview (medium)

GPT-5.5 (Non-reasoning)

K-

K-EXAONE 2.0

GPT-5.6 Sol (Non-reasoning)

Qwen3 VL 32B

DeepSeek V3.1

Grok 3

MiniMax M1 80k

DeepSeek R1 0528

Nova 2.0 Omni (medium)

Gemini 2.5 Flash-Lite

DeepSeek R1 (Jan)

MiniMax M1 40k

Nemotron 3.5 Lightning

K2 Think V2

GLM-4.6

Nova 2.0 Lite (low)

GPT-5.6 Terra (Non-reasoning)

GPT-4o (ChatGPT)

GPT-5.4 (Non-reasoning)

Nova 2.0 Omni (low)

Kimi K2 0905

GLM-5.1

Magistral Medium 1.2

Ring-1T

Kimi K2

Gemini 3 Flash

Grok 3 mini Reasoning (high)

Claude 4.5 Sonnet

Qwen3 Next 80B A3B

GLM-4.5

EXAONE 4.5 33B

Gemini 2.5 Flash-Lite (Sep)

Grok Code Fast 1

gpt-oss-120b (high)

Claude 3.7 Sonnet

Llama 4 Maverick

K-

K-EXAONE

DeepSeek V4 Pro

Command A+

Qwen3.5 Omni Flash

Qwen3 Max

Claude 4.5 Haiku

Gemini 2.5 Flash

Step 3.5 Flash

Mistral Small 4

DeepSeek V3.1

gpt-oss-120b (low)

GLM-4.5-Air

Claude 4 Sonnet

Qwen3 Coder 480B

GPT-4.1 mini

DeepSeek V3.2 Exp

Qwen3.5 9B

GLM-4.6V

DeepSeek V3.1 Terminus

GLM-5.2

GPT-5.1

GPT-5 nano (high)

GLM-5

DeepSeek V3.2

Qwen3 VL 30B A3B

Gemma 4 31B

Qwen3 Max (Preview)

GPT-5 nano (medium)

Qwen3 Coder Next

o3-mini (high)

NVIDIA Nemotron Nano 12B v2 VL

GPT-5.2

DeepSeek V3 0324

Mercury 2

GLM-4.7-Flash

G9v3-3B

Gemma 4 26B A4B

GLM-4.7

Nemotron 3 Nano Omni 30B A3B

Claude 4 Opus

MiMo-V2.5-Pro

GPT-4o (Aug)

Hy3-preview

GPT-5.6 Luna (Non-reasoning)

Nemotron Cascade 2 30B A3B

Trinity Large Thinking

Ling-2.6-1T

Nemotron 3 Nano

Celeris-1

DeepSeek V4 Flash

KAT-Coder-Pro V1

Qwen3 4B 2507

Solar Open 100B

GPT-5 mini (minimal)

Qwen3 VL 8B

INTELLECT-3

Claude 4 Opus

K2-V2 (high)

North Mini Code

Ling-1T

HyperNova 60B 2605

MiMo-V2-Flash

Llama Nemotron Super 49B v1.5

Mistral Large 3

GPT-5.4 mini

Qwen3 VL 32B

Qwen3.5 4B

gpt-oss-20b (high)

Qwen3 235B 2507

Gemma 4 E4B

Gemini 2.5 Flash-Lite

Qwen3 VL 235B A22B

DeepSeek V3 (Dec)

Mistral Medium 3

Devstral Medium

Gemini 2.0 Flash

Qwen3 Coder 30B A3B

Gemma 4 12B (Non-reasoning)

Devstral 2

Nova Premier

Solar Pro 3

gpt-oss-20b (low)

Nova 2.0 Pro Preview

Llama 4 Scout

Devstral Small (May)

Grok 4.1 Fast

Qwen3.5 2B

LongCat Flash Lite

GLM-4.6

K2-V2 (medium)

Grok 4.3 (Non-reasoning)

Ling 2.6 Flash

GPT-5.4 nano

Qwen3 VL 30B A3B

GPT-5 (minimal)

Devstral Small 2

Kimi Linear 48B A3B Instruct

QwQ-32B

NVIDIA Nemotron Nano 9B V2

Claude 3.5 Haiku

Qwen3 30B A3B 2507

Ministral 3 8B

Claude 3 Haiku

Ministral 3 14B

Llama 3.1 405B

Mistral Small 4

Llama Nemotron Super 49B v1.5

Nova 2.0 Omni

Ring-flash-2.0

NVIDIA Nemotron Nano 9B V2

Qwen3 VL 4B

Hermes 4 405B

Nova Pro

Mistral Small 3.1

Grok 4 Fast

Gemma 4 E4B

Cogito v2.1

Grok 4.20 0309

Tri-21B-think Preview

Mistral Medium 3.1

Apriel-v1.5-15B-Thinker

Hermes 4 405B

Grok 4.20 0309 v2

Granite 4.1 30B

GPT-5 nano (minimal)

Command A

Qwen2.5 72B

K2-V2 (low)

NVIDIA Nemotron Nano 12B v2 VL

GPT-4.1 nano

Nova Lite

Mistral Small 3.2

Devstral Small

Nova 2.0 Lite

Jamba 1.7 Large

DiffusionGemma 26B A4B

GLM-4.7-Flash

Magistral Small 1.2

Llama 3.1 8B

Llama 3.3 Nemotron Super 49B

Nemotron 3 Nano 4B

Gemma 4 E2B

Qwen3 VL 8B

Phi-4 Mini

Ministral 3 3B

Ling-flash-2.0

Llama 3.2 11B (Vision)

Llama 3.3 70B

Jamba 1.7 Mini

Qwen3.5 2B

Tri-21B-Think

EXAONE 4.0 32B

DeepSeek R1 0528 Qwen3 8B

Gemma 4 E2B

GLM-4.6V

Qwen3 VL 4B

HyperCLOVA X SEED Think (32B)

Motif-2-12.7B

Llama 3.3 Nemotron Super 49B

Mi:dm K 2.5 Pro Preview

JT-MINI

Pixtral Large

Granite 4.1 8B

Falcon-H1R-7B

Granite 4.0 H Small

Qwen3.5 0.8B

Nova Micro

Qwen3 4B 2507

Mi:dm K 2.5 Pro

DeepSeek R1 Distill Llama 70B

EXAONE 4.0 32B

Qwen3.5 0.8B

Nemotron 3 Nano

DeepSeek R1 Distill Qwen 32B

Jamba Reasoning 3B

Gemma 3 12B

Llama 3.1 70B

Hermes 4 70B

Llama Nemotron Ultra

Llama 3.1 Nemotron 70B

DeepSeek R1 Distill Qwen 14B

ERNIE 5.0 Thinking Preview

Gemma 3 4B

MiniCPM-V 4.6 1.3B

Gemma 3 27B

Ling-mini-2.0

Llama 3.2 1B

Granite 4.0 1B

Granite 4.0 H 1B

Granite 4.0 Micro

Mistral Large 2 (Nov)

LFM2.5-2.6B

MiniCPM5-1B

ERNIE 4.5 300B A47B

Granite 3.3 8B

Granite 4.1 3B

Hermes 4 70B

Phi-3 Mini

Llama 3.2 3B

Mistral Large 2 (Jul)

DeepSeek R1 Distill Qwen 1.5B

LFM2 8B A1B

GLM-4.5V

Gemma 3 270M

Qwen3 Omni 30B A3B

DeepSeek R1 Distill Llama 8B

Qwen3 0.6B

LFM2 1.2B

Magistral Small 1

Qwen3 4B

Mistral 7B

Qwen3 1.7B

GLM-4.5V

Olmo 3 7B Think

Sarvam M

Olmo 3.1 32B Think

LFM2.5-VL-1.6B

Qwen3 30B

Molmo2-8B

Qwen3 235B

Solar Pro 2

Qwen3 14B

Sarvam 105B (high)

LFM2.5-1.2B-Instruct

Apertus 70B Instruct

Qwen3 0.6B

OLMo 2 32B

Granite 4.0 350M

Qwen3 1.7B

LFM2.5-1.2B-Thinking

Magistral Medium 1

LFM2 2.6B

Sarvam 30B (high)

Nanbeige4.1-3B

Tiny Aya Global

Exaone 4.0 1.2B

Llama 3 8B

Apertus 8B Instruct

Olmo 3 7B

Olmo 3.1 32B Instruct

Qwen3 8B

Granite 4.0 H 350M

Gemma 3n E4B

Gemma 3n E2B

Llama 3 70B

Step3 VL 10B

Molmo 7B-D

Qwen3 32B

Llama 3.1 Nemotron Nano 4B v1.1

Qwen3 8B

Qwen3 Omni 30B A3B

Qwen3 235B

Qwen3 32B

GPT-4o (Nov)

Olmo 3 32B Think

Gemma 3 1B

Exaone 4.0 1.2B

Reka Flash 3

LFM2.5-8B-A1B

Mistral Small 3

Solar Pro 2

Qwen3 14B

Phi-4

Qwen3 30B

OLMo 2 7B

LFM2 24B A2B

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.