Code & Software Engineering

Artificial Analysis public leaderboard

Terminal-Bench 2.1

Agentic terminal and software-engineering results published by Artificial Analysis.

Published source

Metric

Accuracy

Results shown

236

Publisher

Artificial Analysis

Snapshot

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

236 of 236 models

Higher is better

Claude Fable 5.1 (max with fallback)

Claude Fable 5.1 (xhigh with fallback)

Claude Fable 5.1 (high with fallback)

GPT-6 Astra (high)

GPT-5.6 Sol (xhigh)

GPT-6 Astra (medium)

GPT-6 Astra (xhigh)

GPT-6 Astra (Non-reasoning)

Claude Opus 5 (max)

GPT-6 Astra (max)

Grok 4.6 (high)

Claude Opus 5 (xhigh)

Claude Fable 5.1 (medium with fallback)

GPT-6 Astra (low)

GPT-5.6 Terra (max)

Grok 4.6 (xhigh)

GPT-5.6 Sol (max)

Claude Opus 5 (high)

Gemini 3.8 Flash (high)

GPT-5.6 Sol (high)

GPT-5.6 Sol (medium)

Qwen3.8-Flash-Next

Claude Opus 5 (medium)

Gemini 3.7 Flash (high)

Muse Spark 1.3 (xhigh)

Claude Fable 5.1 (low with fallback)

Kimi K3 (max)

Claude Opus 4.8 (max)

Claude Fable 5 (with fallback)

GLM-5.3-Flash

GPT-5.5 (xhigh)

Grok 4.6 (medium)

Muse Spark 1.3 (max)

Gemini 3.8 Flash (medium)

GLM-5.3 (max)

Claude Opus 4.7 (max)

Gemini 3.8 Flash (low)

Kimi K3 (low)

Qwen3.8 2.4T A95B

Grok 4.5 (high)

Qwen3.8 Max

GPT-5.6 Luna (max)

Claude Sonnet 5 (max)

GPT-5.5 (medium)

Muse Spark 1.2 (xhigh)

GPT-5.6 Terra (xhigh)

Qwen3.8 27B (xhigh)

Gemini 3.7 Flash (low)

GPT-5.5 (high)

Gemini 3.5 Flash

DeepSeek V4 Pro 0813 (max)

DeepSeek V4 Flash 0731 (max)

GPT-5.4 (xhigh)

Gemini 3.7 Flash (medium)

GPT-5.6 Luna (xhigh)

Muse Spark 1.1 (xhigh)

GLM-5.2 (max)

Gemini 3.6 Flash

GPT-5.6 Sol (low)

Claude Opus 5 (low)

GPT-5.6 Terra (high)

Grok 4.6 (low)

Claude Sonnet 5 (Non-reasoning)

Motif 3

Qwen3.7 Max

DeepSeek V4 Flash Vision (max)

GPT-5.6 Sol (Non-reasoning)

Gemini 3.1 Pro Preview

GPT-5.6 Terra (medium)

K2 Horizon 375B A23B

Claude Sonnet 4.6 (max)

Motif 3 (Beta)

KAT-Coder-Pro V2

Apodex 1.1

GPT-5.6 Luna (high)

Agnes 2.5 Pro Beta

Quasar 438B (max)

Nex-N2-Pro

Qwen3.8 27B (low)

Kimi K2.7 Code

Agnes 2.5 Pro Alpha

Kimi K2.6

GPT-5.5 (low)

MiMo-V2.5-Pro

MiniMax-M3

Qwen3.8 27B (medium)

DeepSeek V4 Pro (high)

Hy3

DeepSeek V4 Pro (max)

MiMo-V2.5

GPT-5.6 Terra (low)

Muse Spark

GLM-5.1

MiMo-V2-Flash (Non-reasoning)

DeepSeek V4 Flash (max)

Qwen3.6 Plus

GPT-5.5 (Non-reasoning)

Qwen3.7 Plus

Qwen3.6 27B

GPT-5.4 nano (xhigh)

JT-4.1 Flash 236B A21B

GPT-5.4 mini (xhigh)

Solar Pro 4

DeepSeek V4 Flash (high)

GPT-5.6 Terra (Non-reasoning)

Claude 4.5 Sonnet

MiniMax-M2.7

Ling 3.0 Flash

Inkling

Inkling Small

Nemotron 3 Ultra

Gemini 3.5 Flash-Lite

GPT-5.6 Luna (medium)

GPT-5.1 (high)

Grok Build 0.1 0616

Muse Glimmer (high)

GLM-5.2 (Non-reasoning)

Qwen3.5 397B A17B

Qwen3.6 27B (Non-reasoning)

Mistral Medium 3.5

LongCat 2.0

GLM-4.6

Qwen3.8 27B

Qwen3.5 122B A10B

Qwen3.5 122B A10B (Non-reasoning)

DeepSeek V3.2

Kimi K2.5

GLM-4.7

DeepSeek V3.1 Terminus

Qwen3.6 35B A3B

Solar Open2 250B

Claude 4.5 Haiku

GPT-5.6 Luna (low)

Gemma 4 31B

Ring-2.6-1T

Qwen3.6 35B A3B (Non-reasoning)

Qwen3.5 35B A3B (Non-reasoning)

K-

K-EXAONE 2.0

Grok 4.3 (high)

Step 3.7 Flash

Gemma 4 26B A4B

A.X-K2

GPT-5.6 Luna (Non-reasoning)

Nemotron 3 Super

Qwen3 Coder Next

Claude 4 Sonnet

North Mini Code

GPT-5 (high)

GPT-5.5 Instant (June 2026)

Grok 4.3 (Non-reasoning)

G9v3-39A5B

Gemini 3.1 Flash-Lite

Devstral 2

K-

K-EXAONE

Nova 2.0 Pro Preview (medium)

Devstral Small 2

Qwen3.5 9B

Gemma 4 31B (Non-reasoning)

Gemini 2.5 Pro

Ling 3.0 Tiny

Mercury 2

Gemma 4 12B

Granite 4.2 30B

Mistral Small 3.1

gpt-oss-120b (high)

Qwen3.5 4B

Nemotron 3.5 Lightning

Ling 2.6 Flash

Command A+

Qwen3.5 9B (Non-reasoning)

Qwen3.5 4B (Non-reasoning)

EXAONE 4.5 33B

Mistral Small 4

Nemotron Cascade 2 30B A3B

Trinity Large Thinking

Nova 2.0 Pro Preview (low)

DeepSeek R1 (Jan)

HyperNova 60B 2605 (high)

Granite 4.2 8B

Nova 2.0 Pro Preview (Non-reasoning)

DeepSeek V3 (Dec)

Nova 2.0 Lite (high)

K2 Think V2

Mistral Medium 3.1

Granite 4.2 3B

gpt-oss-20b (high)

DeepSeek V3 0324

gpt-oss-120b (low)

Magistral Medium 1.2

DiffusionGemma 26B A4B

Mistral Large 3

Solar Pro 3

Qwen3 235B A22B 2507

Celeris-1

Claude 3.5 Haiku

GPT-4.1 mini

Ministral 3 14B

MiniCPM5-2B

Llama 4 Maverick

Nemotron 3 Nano

Nemotron 3 Nano Omni 30B A3B

Qwen3 Next 80B A3B (Reasoning)

G9v3-3B

Mistral Small 3.2

GPT-4o mini

Qwen3 32B

Qwen3 14B

Llama 3.3 70B

o3-mini (high)

Gemma 3 27B

Magistral Small 1.2

LFM2.5-2.6B

Ministral 3 8B

GPT-5 mini (high)

GPT-4.1 nano

Nemotron 3 Nano 4B

Llama 4 Scout

Granite 4.1 8B

Qwen3.5 2B

Granite 4.1 30B

Qwen3 8B

Gemma 4 E4B

Qwen3 30B A3B 2507

Llama 3.1 8B

Nanbeige4.1-3B

Granite 4.1 3B

Gemma 3n E4B

Gemma 3 4B

Qwen3.5 0.8B (Non-reasoning)

Phi-4 Mini

Gemma 4 E2B

Qwen3.5 2B (Non-reasoning)

Gemma 3 12B

MiniCPM-V 4.6 1.3B

Ministral 3 3B

Qwen3.5 0.8B

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.