Code & Software Engineering

Artificial Analysis public leaderboard

SciCode

Scientific programming benchmark results published by Artificial Analysis.

Published source

Metric

Accuracy

Results shown

575

Publisher

Artificial Analysis

Snapshot

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

575 of 575 models

Higher is better

Claude Fable 5 (with fallback)

Gemini 3.1 Pro Preview

Kimi K3 (max)

Muse Spark 1.1 (xhigh)

Gemini 3.7 Flash (medium)

GPT-5.6 Sol (high)

Gemini 3.7 Flash (high)

GPT-5.4 (xhigh)

GPT-5.6 Sol (medium)

GLM-5.3 (max)

Muse Spark 1.2 (xhigh)

GPT-5.5 (xhigh)

Gemini 3 Pro Preview (high)

GPT-5.6 Sol (max)

GPT-5.6 Sol (xhigh)

GPT-5.5 (high)

Claude Opus 5 (max)

GPT-5.6 Sol (low)

Claude Opus 5 (xhigh)

GPT-5.2 Codex (xhigh)

Grok 4.6 (medium)

Claude Opus 4.7 (max)

Claude Opus 5 (high)

Grok 4.5 (high)

GPT-5.6 Terra (max)

Claude Sonnet 5 (max)

Grok 4.6 (high)

Gemini 3.7 Flash (low)

Kimi K2.6

GPT-5.5 (medium)

Claude Opus 4.8 (max)

GPT-5.3 Codex (xhigh)

Gemini 3.5 Flash

Gemini 3.5 Flash (medium)

Qwen3.8 Max

Gemini 3.6 Flash

GPT-5.6 Luna (max)

GPT-5.2 (xhigh)

Claude Opus 4.6 (max)

Qwen3.8 2.4T A95B

GPT-5.6 Terra (xhigh)

GPT-5.5 (low)

Grok 4.6 (xhigh)

Muse Spark

Kimi K3 (low)

GPT-5.6 Luna (high)

Claude Opus 5 (medium)

Gemini 3 Flash

GLM-5.2 (max)

GPT-5.4 (low)

GPT-5.5 Instant (May 2026)

MiMo-V2.5-Pro

Grok Build 0.1 0616

Claude Opus 4.7 (Non-reasoning, high)

GPT-5.6 Terra (high)

GPT-5.6 Luna (xhigh)

DeepSeek V4 Pro (max)

Gemini 3 Flash

Gemini 3 Pro Preview (low)

GPT-5.4 mini (xhigh)

DeepSeek V4 Flash 0731 (max)

GPT-5.6 Terra (medium)

Claude Opus 4.5

DeepSeek V4 Pro 0813 (max)

GPT-5.6 Terra (low)

Kimi K2.5

Gemini 3.5 Flash (minimal)

Qwen3.7 Max

Inkling Small

GPT-5.5 Instant (June 2026)

Claude Sonnet 5 (Non-reasoning)

Grok 4.6 (low)

Claude Opus 5 (low)

Hy3

Kimi K2.7 Code

GPT-5.5 (Non-reasoning)

Grok 4.3 (high)

GPT-5.4 (Non-reasoning)

GPT-5.6 Sol (Non-reasoning)

Claude Opus 4.5

MiniMax-M2.7

Claude Sonnet 4.6 (Non-reasoning)

Qwen3.6 Max Preview

GPT-5.4 nano (xhigh)

Claude Sonnet 4.6 (max)

o4-mini (high)

DeepSeek V4 Pro (high)

GLM-5

GPT-5.2 (medium)

Inkling

GPT-5.6 Luna (medium)

Claude Opus 4.6 (high)

Grok 4

GPT-5.6 Luna (low)

Solar Open2 250B

Grok 4.20 0309 v2

Qwen3.7 Plus

MiniMax-M3

GLM-4.7

DeepSeek V4 Flash (max)

Grok 4.20 0309

Claude 4.5 Sonnet

Qwen3.8 27B (xhigh)

Grok 4.3 (medium)

GPT-5.6 Terra (Non-reasoning)

Motif 3 (Beta)

Grok 4.1 Fast

GPT-5.4 mini (medium)

Grok 4 Fast

Claude Sonnet 4.6 (Non-reasoning, Low Effort)

DeepSeek V3.2 Speciale

GLM-5.1

GLM-5-Turbo

Solar Pro 4

Muse Glimmer (high)

GLM 5V Turbo

Gemma 4 31B

GPT-5.1 (high)

Claude 4.5 Haiku

MiMo-V2.5

Qwen3 Max Thinking

GPT-5 (high)

Gemini 2.5 Pro

Claude 4.5 Sonnet

Nova 2.0 Pro Preview (medium)

MiniMax-M2.5

GPT-5.1 Codex mini (high)

MiMo-V2-Pro

DeepSeek V4 Pro

Kimi K2 Thinking

Ring-2.6-1T

Qwen3 235B A22B 2507

Agnes 2.5 Pro Alpha

Qwen3.5 397B A17B

Qwen3.5 122B A10B

DeepSeek V4 Flash (high)

Gemini 3.1 Flash-Lite

Grok 4.3 (low)

Nex-N2-Pro

Gemini 2.5 Pro (May)

Hy3-preview

Qwen3.5 397B A17B

GPT-5 (medium)

Gemma 4 31B

Ling 3.0 Flash

K-

K-EXAONE 2.0

GPT-5 mini (medium)

Cogito v2.1

Claude 4.1 Opus

Gemini 3.5 Flash-Lite

Claude 4 Opus

GPT-5 Codex (high)

MiniMax-M2.1

Doubao Seed Code

Qwen3.6 Plus

DeepSeek V3.1 Terminus

Motif 3

Grok 3 mini Reasoning (high)

Qwen3.5 Omni Plus

Gemini 2.5 Flash (Sep)

GPT-5.2

GPT-4.1 mini

Step 3.5 Flash

Claude 3.7 Sonnet

DeepSeek R1 0528

GPT-5.1 Codex (high)

Step 3.7 Flash

Gemma 4 26B A4B

Claude 4 Sonnet

DeepSeek V3.2 Exp

o3-mini

Nemotron 3 Ultra

Qwen3 VL 235B A22B

Qwen3 235B

GPT-5.6 Luna (Non-reasoning)

Qwen3.8 27B (low)

Claude 4 Opus

Qwen3.6 27B

o3-mini (high)

GPT-5.4 mini

Kimi K2.5

Mistral Medium 3.5

Kimi K2.6

Qwen3.5 27B

MiMo-V2-Omni-0327

Gemini 2.5 Pro (Mar)

Gemini 2.5 Flash

Hy3-preview

MiMo-V2-Flash

GPT-5 mini (high)

Magistral Medium 1.2

DeepSeek V3.1

INTELLECT-3

MiMo-V2.5-Pro

GPT-5 (low)

DeepSeek V3.2

gpt-oss-120b (high)

GPT-5 (minimal)

Qwen3 Next 80B A3B

Mercury 2

DeepSeek V3.2

Qwen3 Max Thinking (Preview)

Nova 2.0 Pro Preview (low)

Step 3.5 Flash 2603

A.X-K2

GLM-4.6

GPT-5.4 nano

MiMo-V2-Flash (Feb 2026)

Qwen3 Max

GLM-5

KAT-Coder-Pro V2

North Mini Code

JT-4.1 Flash 236B A21B

Gemma 4 12B

GPT-4.1

Qwen3.8 27B (medium)

Mistral Small 4

Command A+

MiniMax M1 40k

GPT-5 (ChatGPT)

Qwen3.5 35B A3B

DeepSeek V3.2 Exp

Claude 3.7 Sonnet

DeepSeek R1 Distill Qwen 32B

Gemini 2.5 Flash (Sep)

ERNIE 5.0 Thinking Preview

Grok 4.3 (Non-reasoning)

MiniMax M1 80k

Qwen3.6 27B

Gemma 4 26B A4B

Apriel-v1.6-15B-Thinker

DeepSeek V4 Flash

Claude 4 Sonnet

Ling-2.6-1T

Qwen3 Max (Preview)

Nova 2.0 Lite (high)

GPT-5 mini (minimal)

Grok 3

Nova 2.0 Lite (medium)

Qwen3.5 27B

MiMo-V2-Omni

Ring-1T

DeepSeek V3.1

Claude 3.5 Sonnet (Oct)

GPT-4o (Mar)

GPT-5 nano (high)

KAT-Coder-Pro V1

Seed-OSS-36B-Instruct

GPT-5.1

Nova 2.0 Omni (medium)

Mistral Large 3

Grok Code Fast 1

GLM-5.1

Trinity Large Thinking

GLM-5.2

MiniMax-M2

Qwen3 235B 2507

Nemotron 3 Super

gpt-oss-120b (low)

Qwen3 Coder 480B

Gemini 2.5 Flash

Qwen3 VL 235B A22B

DeepSeek V3 0324

Qwen3.6 35B A3B

QwQ-32B

DeepSeek R1 (Jan)

Qwen3.5 122B A10B

K-

K-EXAONE

DeepSeek V3 (Dec)

LongCat 2.0

GLM-4.7

Qwen3 32B

Ling-1T

Magistral Small 1.2

GPT-5.4 nano

GLM-4.5

Nemotron Cascade 2 30B A3B

Apriel-v1.5-15B-Thinker

Llama Nemotron Super 49B v1.5

Llama Nemotron Ultra

Hermes 4 405B

Kimi K2

gpt-oss-20b (high)

EXAONE 4.0 32B

Claude 4.5 Haiku

Nova 2.0 Omni (low)

DiffusionGemma 26B A4B

Hermes 4 70B

gpt-oss-20b (low)

G9v3-39A5B

Gemini 2.0 Flash (exp)

Mistral Medium 3.1

GPT-5 nano (medium)

GLM-4.7-Flash

Qwen2.5 Max

GPT-4o (ChatGPT)

Qwen3 30B A3B 2507

Nova 2.0 Lite (low)

Gemini 2.0 Flash

GPT-4o (Nov)

Mi:dm K 2.5 Pro

Devstral 2

Mistral Medium 3

GPT-4o (Aug)

Llama 4 Maverick

GLM-4.6

HyperNova 60B 2605

K2 Think V2

Grok 4 Fast

Gemini 2.0 Flash Thinking exp. (Jan)

Grok 4.20 0309 v2

o1-mini

Qwen3 Coder Next

Grok 4.20 0309

DeepSeek V3.1 Terminus

GPT-4 Turbo

Nemotron 3.5 Lightning

Qwen3 14B

Claude 3.5 Sonnet (June)

ERNIE 4.5 300B A47B

DeepSeek R1 Distill Llama 70B

Gemini 2.0 Pro Experimental

Step3 VL 10B

GPT-4o (May)

Qwen3 VL 30B A3B

Qwen3 Next 80B A3B

Kimi K2 0905

GLM-4.5-Air

Qwen3 Omni 30B A3B

Qwen3 30B A3B 2507

GLM-4.6V

Solar Pro 2

Tulu3 405B

Qwen3 VL 32B

Qwen3 235B

Llama 3.1 405B

Gemma 4 12B (Non-reasoning)

Mi:dm K 2.5 Pro Preview

Magistral Medium 1

Grok 4.1 Fast

Nemotron 3 Nano

Gemini 1.5 Pro (Sep)

Grok Beta

Devstral Medium

Olmo 3.1 32B Think

Qwen3.5 35B A3B

Mistral Large 2 (Nov)

Pixtral Large

JT-35B-Flash

GPT-5 nano (minimal)

Gemini 2.5 Flash

Ling-flash-2.0

Qwen3 VL 30B A3B

Devstral Small 2

Gemini 2.5 Flash-Lite (Sep)

K2-V2 (high)

Olmo 3 32B Think

Grok 2

Qwen3 30B

Gemini 2.5 Flash-Lite (Sep)

Qwen3 VL 32B

HyperCLOVA X SEED Think (32B)

LongCat Flash Lite

Llama 3.3 Nemotron Super 49B

Motif-2-12.7B

Mistral Small 4

Command A

Nova 2.0 Pro Preview

EXAONE 4.5 33B

Qwen3 32B

Nova 2.0 Omni

Nova Premier

Nemotron 3 Nano Omni 30B A3B

Qwen3 Coder 30B A3B

Qwen3.5 9B

Hermes 4 70B

Qwen3.5 9B

Claude 3.5 Haiku

Gemini 1.5 Pro (May)

JT-MINI

GLM-4.6V

Solar Pro 2

Qwen2.5 Coder 32B

Mistral Large 2 (Jul)

Ling 2.6 Flash

K-

K-EXAONE

Solar Open 100B

Gemini 1.5 Flash (Sep)

Llama 3.1 70B

Qwen2.5 72B

Reka Flash 3

Nanbeige4.1-3B

Mistral Small 3.1

Qwen3 14B

Mistral Small 3.2

Sarvam 105B (high)

Qwen3 30B

NVIDIA Nemotron Nano 12B v2 VL

Llama 3.3 70B

Phi-4

GPT-4.1 nano

MiMo-V2-Flash

Granite 4.1 30B

Qwen3 4B 2507

Qwen3.5 Omni Flash

GLM-4.7-Flash

EXAONE 4.0 32B

Hermes 4 405B

K2-V2 (medium)

Gemini 2.0 Flash-Lite (Feb)

Falcon-H1R-7B

Solar Pro 2

Gemini 2.0 Flash-Lite (Preview)

Solar Pro 3

Devstral Small (May)

Gemma 4 E4B

Devstral Small

Ling 3.0 Tiny

Magistral Small 1

Mistral Saba

Nova 2.0 Lite

Llama 3.2 90B (Vision)

DeepSeek R1 Distill Qwen 14B

Llama Nemotron Super 49B v1.5

Ministral 3 14B

Mistral Small 3

Gemini 2.5 Flash

Llama 3.1 Nemotron 70B

Claude 3 Opus

Hermes 3 - Llama-3.1 70B

Nemotron 3 Nano

Qwen2.5 Instruct 32B

Sonar

Qwen2 72B

Gemini 1.5 Flash-8B

Claude 3 Sonnet

GPT-4o mini

Llama 3.3 Nemotron Super 49B

DeepHermes 3 - Mistral 24B

Sonar Pro

Qwen3 8B

K2-V2 (low)

GLM-4.5V

NVIDIA Nemotron Nano 9B V2

Qwen3 VL 8B

Granite 4.1 8B

Gemma 3 27B

Olmo 3 7B Think

NVIDIA Nemotron Nano 9B V2

Granite 4.0 H Small

Gemma 4 E2B

Nova Pro

Mistral Large (Feb)

Ministral 3 8B

Celeris-1

DeepSeek R1 0528 Qwen3 8B

Gemma 4 E2B

Kimi Linear 48B A3B Instruct

Claude 2.0

Gemini 2.5 Flash-Lite

Sarvam 30B (high)

Llama 3 70B

GLM-4.5V

Jamba 1.7 Large

Mixtral 8x22B

Qwen3 Omni 30B A3B

Claude 3 Haiku

Jamba 1.6 Large

Claude 2.1

Qwen3.5 4B

Gemini 1.5 Flash (May)

Qwen3 4B 2507

Sarvam M

Tri-21B-think Preview

Gemini 2.5 Flash-Lite

G9v3-3B

NVIDIA Nemotron Nano 12B v2 VL

Gemma 3 12B

Qwen3 VL 8B

Tri-21B-Think

Qwen3 VL 4B

Llama 4 Scout

Ring-flash-2.0

Qwen3 8B

Olmo 3.1 32B Instruct

Qwen3 4B

Solar Pro 2

Nemotron 3 Nano 4B

Jamba 1.5 Large

Qwen3.5 4B

Mistral Small (Sep)

Qwen2.5 Turbo

Qwen2.5 Coder 7B

Ministral 3 3B

LFM2.5-2.6B

Nova Lite

DeepSeek Coder V2 Lite

Qwen3 VL 4B

Ling-mini-2.0

Mistral Small (Feb)

Molmo2-8B

Llama 3.1 8B

Llama 3 8B

Granite 4.0 Micro

Granite 4.1 3B

DeepSeek R1 Distill Llama 8B

DBRX

Mistral Medium

Llama 2 Chat 13B

Command-R+ (Apr)

Gemini 1.0 Pro

Llama 3.2 11B (Vision)

Phi-4 Multimodal

LFM2 24B A2B

Phi-4 Mini

Olmo 3 7B

Granite 3.3 8B

Llama 3.1 Nemotron Nano 4B v1.1

Jamba 1.6 Mini

Nova Micro

Jamba 1.7 Mini

Exaone 4.0 1.2B

DeepHermes 3 - Llama-3.1 8B

Phi-3 Mini

Granite 4.0 1B

Gemma 3n E4B (May)

Granite 4.0 H 1B

Gemma 3n E4B

Jamba 1.5 Mini

OLMo 2 32B

LFM2.5-8B-A1B

Exaone 4.0 1.2B

Gemma 3 4B

Qwen3.5 2B

LFM 40B

Qwen3 1.7B

LFM2 8B A1B

DeepSeek R1 Distill Qwen 1.5B

Command-R (Mar)

Jamba Reasoning 3B

Apertus 70B Instruct

Llama 3.2 3B

Gemma 3n E2B

MiniCPM5-1B

Qwen3 1.7B

LFM2.5-1.2B-Thinking

Qwen3 0.6B

Apertus 8B Instruct

Gemma 4 E4B

QwQ 32B-Preview

OLMo 2 7B

Tiny Aya Global

Molmo 7B-D

Qwen3 4B

LFM2.5-VL-1.6B

Qwen3.5 0.8B

Mixtral 8x7B

Qwen3 0.6B

Qwen3.5 2B

LFM2 1.2B

LFM2 2.6B

Mistral 7B

LFM2.5-1.2B-Instruct

MiniCPM-V 4.6 1.3B

Llama 3.2 1B

Granite 4.0 H 350M

MiniCPM5-1B

Qwen3.6 35B A3B

Granite 4.0 350M

Gemma 3 1B

Gemma 3 270M

Qwen3.5 0.8B

Llama 2 Chat 7B

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.