How It Works Ecosystem

The open directory of AI agents, LLMs, MCP servers, workflows, and tools — search, compare, and track status across platforms.

Explore

Browse All AI Agents LLMs MCP Servers Tools & Frameworks

Account

Sign In Create Account Submit an Agent Saved Agents

Support

How It Works FAQ Report an Issue

Company

ATMEZ AI Solutions

© 2026 ATMEZ AI Solutions · All rights reserved.

Built with ❤️ for the AI community

All Agents LLMs MCP Servers Workflows Frameworks Tools Skills GenAI News

Terminal-Bench 2.1 Leaderboard | Genaihub

Code & Software Engineering

Artificial Analysis public leaderboard

Terminal-Bench 2.1

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

Published source(opens in a new tab)

Metric

Accuracy

Results shown

208

Publisher

Artificial Analysis

Snapshot

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

200 of 200 models

A.

A.X-K2

Ag

Agnes 2.5 Pro Alpha

Ce

Celeris-1

Claude 3.5 Haiku

Claude 4 Sonnet

Claude 4.5 Haiku

Claude 4.5 Sonnet

Claude Fable 5 (with fallback)

Claude Opus 4.7 (max)

Claude Opus 4.8 (max)

Claude Opus 5 (high)

Claude Opus 5 (low)

Claude Opus 5 (max)

Claude Opus 5 (medium)

Claude Opus 5 (xhigh)

Claude Sonnet 4.6 (max)

Claude Sonnet 5 (max)

Claude Sonnet 5 (Non-reasoning)

Co

Command A+

DeepSeek R1 (Jan)

DeepSeek V3 (Dec)

DeepSeek V3 0324

DeepSeek V3.1 Terminus

DeepSeek V3.2

DeepSeek V4 Flash (high)

DeepSeek V4 Flash (max)

DeepSeek V4 Flash 0731 (max)

DeepSeek V4 Pro (high)

DeepSeek V4 Pro (max)

DeepSeek V4 Pro 0813 (max)

De

Devstral 2

De

Devstral Small 2

Di

DiffusionGemma 26B A4B

EX

EXAONE 4.5 33B

G9

G9v3-39A5B

G9

G9v3-3B

Gemini 2.5 Pro

Gemini 3.1 Flash-Lite

Gemini 3.1 Pro Preview

Gemini 3.5 Flash

Gemini 3.5 Flash-Lite

Gemini 3.6 Flash

Gemini 3.7 Flash (high)

Gemini 3.7 Flash (low)

Gemini 3.7 Flash (medium)

Ge

Gemma 3 12B

Ge

Gemma 3 27B

Ge

Gemma 3 4B

Ge

Gemma 3n E4B

Ge

Gemma 4 12B

Ge

Gemma 4 26B A4B

Ge

Gemma 4 31B

Ge

Gemma 4 E2B

Ge

Gemma 4 E4B

GLM-4.6

GLM-4.7

GLM-5.1

GLM-5.2

GLM-5.2 (max)

GLM-5.3 (max)

GPT-4.1 mini

GPT-4.1 nano

GPT-4o mini

GPT-5 (high)

GPT-5 mini (high)

GPT-5.1 (high)

GPT-5.4 (xhigh)

GPT-5.4 mini (xhigh)

GPT-5.4 nano (xhigh)

GPT-5.5 (high)

GPT-5.5 (low)

GPT-5.5 (medium)

GPT-5.5 (Non-reasoning)

GPT-5.5 (xhigh)

GPT-5.5 Instant (June 2026)

GPT-5.6 Luna (high)

GPT-5.6 Luna (low)

GPT-5.6 Luna (max)

GPT-5.6 Luna (medium)

GPT-5.6 Luna (Non-reasoning)

GPT-5.6 Luna (xhigh)

GPT-5.6 Sol (high)

GPT-5.6 Sol (low)

GPT-5.6 Sol (max)

GPT-5.6 Sol (medium)

GPT-5.6 Sol (Non-reasoning)

GPT-5.6 Sol (xhigh)

GPT-5.6 Terra (high)

GPT-5.6 Terra (low)

GPT-5.6 Terra (max)

GPT-5.6 Terra (medium)

GPT-5.6 Terra (Non-reasoning)

GPT-5.6 Terra (xhigh)

gpt-oss-120b (high)

gpt-oss-120b (low)

gpt-oss-20b (high)

Gr

Granite 4.1 30B

Gr

Granite 4.1 3B

Gr

Granite 4.1 8B

Grok 4.3 (high)

Grok 4.3 (Non-reasoning)

Grok 4.5 (high)

Grok 4.6 (high)

Grok 4.6 (low)

Grok 4.6 (medium)

Grok 4.6 (xhigh)

Grok Build 0.1 0616

Hy

Hy3

Hy

HyperNova 60B 2605

In

Inkling

In

Inkling Small

JT

JT-4.1 Flash 236B A21B

K-

K-EXAONE

K-

K-EXAONE 2.0

K2

K2 Think V2

KA

KAT-Coder-Pro V2

Kimi K2.5

Kimi K2.6

Kimi K2.7 Code

Kimi K3 (low)

Kimi K3 (max)

LF

LFM2.5-2.6B

Li

Ling 2.6 Flash

Li

Ling 3.0 Flash

Li

Ling 3.0 Tiny

Llama 3.1 8B

Llama 3.3 70B

Llama 4 Maverick

Llama 4 Scout

Lo

LongCat 2.0

Ma

Magistral Medium 1.2

Ma

Magistral Small 1.2

Me

Mercury 2

MiMo-V2-Flash

MiMo-V2.5

MiMo-V2.5-Pro

Mi

MiniCPM-V 4.6 1.3B

MiniMax-M2.7

MiniMax-M3

Mi

Ministral 3 14B

Mi

Ministral 3 3B

Mi

Ministral 3 8B

Mi

Mistral Large 3

Mi

Mistral Medium 3.1

Mi

Mistral Medium 3.5

Mi

Mistral Small 3.1

Mi

Mistral Small 3.2

Mi

Mistral Small 4

Mo

Motif 3

Mo

Motif 3 (Beta)

Muse Glimmer (high)

Muse Spark

Muse Spark 1.1 (xhigh)

Muse Spark 1.2 (xhigh)

Na

Nanbeige4.1-3B

Ne

Nemotron 3 Nano

Ne

Nemotron 3 Nano 4B

Ne

Nemotron 3 Nano Omni 30B A3B

Ne

Nemotron 3 Super

Ne

Nemotron 3 Ultra

Ne

Nemotron 3.5 Lightning

Ne

Nemotron Cascade 2 30B A3B

Ne

Nex-N2-Pro

No

North Mini Code

No

Nova 2.0 Lite (high)

No

Nova 2.0 Pro Preview

No

Nova 2.0 Pro Preview (low)

No

Nova 2.0 Pro Preview (medium)

o3-mini (high)

Ph

Phi-4 Mini

Qwen3 14B

Qwen3 235B A22B 2507

Qwen3 30B A3B 2507

Qwen3 32B

Qwen3 8B

Qwen3 Coder Next

Qwen3 Next 80B A3B

Qwen3.5 0.8B

Qwen3.5 122B A10B

Qwen3.5 2B

Qwen3.5 35B A3B

Qwen3.5 397B A17B

Qwen3.5 4B

Qwen3.5 9B

Qwen3.6 27B

Qwen3.6 35B A3B

Qwen3.6 Plus

Qwen3.7 Max

Qwen3.7 Plus

Qwen3.8 2.4T A95B

Qwen3.8 27B (low)

Qwen3.8 27B (medium)

Qwen3.8 27B (xhigh)

Qwen3.8 Max

Ri

Ring-2.6-1T

So

Solar Open2 250B

So

Solar Pro 3

So

Solar Pro 4

St

Step 3.7 Flash

Tr

Trinity Large Thinking

208 of 208 models

Higher is better

25

50

75

100

90

GPT-5.6 Sol (xhigh)

89

Claude Opus 5 (max)

88

Grok 4.6 (high)

88

Claude Opus 5 (xhigh)

88

GPT-5.6 Terra (max)

88

Grok 4.6 (xhigh)

88

GPT-5.6 Sol (max)

88

Claude Opus 5 (high)

87

GPT-5.6 Sol (high)

86

GPT-5.6 Sol (medium)

86

Claude Opus 5 (medium)

86

Gemini 3.7 Flash (high)

85

Kimi K3 (max)

85

Claude Opus 4.8 (max)

85

Claude Fable 5 (with fallback)

84

GPT-5.5 (xhigh)

84

Grok 4.6 (medium)

84

GLM-5.3 (max)

83

Claude Opus 4.7 (max)

82

Kimi K3 (low)

82

Qwen3.8 2.4T A95B

82

Grok 4.5 (high)

81

Qwen3.8 Max

81

GPT-5.6 Luna (max)

81

Claude Sonnet 5 (max)

81

GPT-5.5 (medium)

80

Muse Spark 1.2 (xhigh)

80

GPT-5.6 Terra (xhigh)

80

Qwen3.8 27B (xhigh)

80

Gemini 3.7 Flash (low)

79

GPT-5.5 (high)

79

Gemini 3.5 Flash

79

DeepSeek V4 Pro 0813 (max)

79

DeepSeek V4 Flash 0731 (max)

78

GPT-5.4 (xhigh)

78

Gemini 3.7 Flash (medium)

78

GPT-5.6 Luna (xhigh)

78

Muse Spark 1.1 (xhigh)

78

GLM-5.2 (max)

78

Gemini 3.6 Flash

77

GPT-5.6 Sol (low)

76

Claude Opus 5 (low)

76

GPT-5.6 Terra (high)

75

Grok 4.6 (low)

75

Claude Sonnet 5 (Non-reasoning)

75

Mo

Motif 3

75

Qwen3.7 Max

74

GPT-5.6 Sol (Non-reasoning)

74

Gemini 3.1 Pro Preview

72

GPT-5.6 Terra (medium)

71

Claude Sonnet 4.6 (max)

71

Mo

Motif 3 (Beta)

70

KA

KAT-Coder-Pro V2

70

GPT-5.6 Luna (high)

68

Ne

Nex-N2-Pro

67

Qwen3.8 27B (low)

67

Kimi K2.7 Code

67

Ag

Agnes 2.5 Pro Alpha

66

Kimi K2.6

66

GPT-5.5 (low)

65

MiMo-V2.5-Pro

65

MiniMax-M3

65

Qwen3.8 27B (medium)

65

DeepSeek V4 Pro (high)

64

Hy

Hy3

64

DeepSeek V4 Pro (max)

64

MiMo-V2.5

63

GPT-5.6 Terra (low)

62

Muse Spark

62

GLM-5.1

62

MiMo-V2-Flash

62

DeepSeek V4 Flash (max)

61

Qwen3.6 Plus

61

GPT-5.5 (Non-reasoning)

61

Qwen3.7 Plus

61

Qwen3.6 27B

61

GPT-5.4 nano (xhigh)

60

JT

JT-4.1 Flash 236B A21B

59

GPT-5.4 mini (xhigh)

57

So

Solar Pro 4

57

DeepSeek V4 Flash (high)

56

GPT-5.6 Terra (Non-reasoning)

56

Claude 4.5 Sonnet

55

MiniMax-M2.7

55

Li

Ling 3.0 Flash

55

In

Inkling

55

In

Inkling Small

54

Ne

Nemotron 3 Ultra

54

Gemini 3.5 Flash-Lite

53

GPT-5.6 Luna (medium)

52

GPT-5.1 (high)

52

Grok Build 0.1 0616

52

Muse Glimmer (high)

52

GLM-5.2

51

Qwen3.5 397B A17B

51

Qwen3.6 27B

51

Mi

Mistral Medium 3.5

50

Lo

LongCat 2.0

49

GLM-4.6

48

Qwen3.5 122B A10B

47

Qwen3.5 122B A10B

47

DeepSeek V3.2

46

Kimi K2.5

45

GLM-4.7

45

DeepSeek V3.1 Terminus

45

Qwen3.6 35B A3B

44

So

Solar Open2 250B

44

Claude 4.5 Haiku

43

GPT-5.6 Luna (low)

43

Ge

Gemma 4 31B

43

Ri

Ring-2.6-1T

42

Qwen3.6 35B A3B

41

Qwen3.5 35B A3B

40

K-

K-EXAONE 2.0

40

Grok 4.3 (high)

39

St

Step 3.7 Flash

39

Ge

Gemma 4 26B A4B

39

A.

A.X-K2

39

GPT-5.6 Luna (Non-reasoning)

39

Ne

Nemotron 3 Super

38

Qwen3 Coder Next

36

Claude 4 Sonnet

36

No

North Mini Code

35

GPT-5 (high)

35

GPT-5.5 Instant (June 2026)

34

Grok 4.3 (Non-reasoning)

33

G9

G9v3-39A5B

31

Gemini 3.1 Flash-Lite

30

De

Devstral 2

30

K-

K-EXAONE

30

No

Nova 2.0 Pro Preview (medium)

30

De

Devstral Small 2

29

Qwen3.5 9B

29

Ge

Gemma 4 31B

28

Gemini 2.5 Pro

28

Li

Ling 3.0 Tiny

27

Me

Mercury 2

27

Ge

Gemma 4 12B

26

Mi

Mistral Small 3.1

26

gpt-oss-120b (high)

26

Qwen3.5 4B

24

Ne

Nemotron 3.5 Lightning

24

Li

Ling 2.6 Flash

23

Co

Command A+

21

Qwen3.5 9B

21

Qwen3.5 4B

21

EX

EXAONE 4.5 33B

21

Mi

Mistral Small 4

21

Ne

Nemotron Cascade 2 30B A3B

21

Tr

Trinity Large Thinking

19

No

Nova 2.0 Pro Preview (low)

19

DeepSeek R1 (Jan)

18

Hy

HyperNova 60B 2605

17

No

Nova 2.0 Pro Preview

17

DeepSeek V3 (Dec)

16

No

Nova 2.0 Lite (high)

15

K2

K2 Think V2

14

Mi

Mistral Medium 3.1

14

gpt-oss-20b (high)

14

DeepSeek V3 0324

14

gpt-oss-120b (low)

12

Ma

Magistral Medium 1.2

12

Di

DiffusionGemma 26B A4B

12

Mi

Mistral Large 3

12

So

Solar Pro 3

12

Qwen3 235B A22B 2507

11

Ce

Celeris-1

10

Claude 3.5 Haiku

10

GPT-4.1 mini

10

Mi

Ministral 3 14B

8

Llama 4 Maverick

7

Ne

Nemotron 3 Nano

7

Ne

Nemotron 3 Nano Omni 30B A3B

7

Qwen3 Next 80B A3B

6

G9

G9v3-3B

6

Mi

Mistral Small 3.2

6

GPT-4o mini

5

Qwen3 32B

5

Qwen3 14B

5

Llama 3.3 70B

4

o3-mini (high)

4

Ge

Gemma 3 27B

4

Ma

Magistral Small 1.2

4

LF

LFM2.5-2.6B

4

Mi

Ministral 3 8B

4

GPT-5 mini (high)

4

GPT-4.1 nano

4

Ne

Nemotron 3 Nano 4B

4

Llama 4 Scout

3

Gr

Granite 4.1 8B

3

Qwen3.5 2B

3

Gr

Granite 4.1 30B

2

Qwen3 8B

2

Ge

Gemma 4 E4B

1

Qwen3 30B A3B 2507

1

Llama 3.1 8B

1

Na

Nanbeige4.1-3B

1

Gr

Granite 4.1 3B

1

Ge

Gemma 3n E4B

0

Ge

Gemma 3 4B

0

Qwen3.5 0.8B

0

Ph

Phi-4 Mini

0

Ge

Gemma 4 E2B

0

Qwen3.5 2B

0

Ge

Gemma 3 12B

0

Mi

MiniCPM-V 4.6 1.3B

0

Mi

Ministral 3 3B

0

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.