Providers
OpenAI

United States

OpenAI

OpenAI develops GPT, DALL·E, and multimodal foundation models, offering APIs and products like ChatGPT for consumers and enterprises.

Products
1
Models
86
Available
0
Benchmarks
17

Region

United States

Updated

Jul 10, 2026

Product coverage

Products from this provider

1

Model coverage

Models from this provider

86

GPT-3.5 Turbo

GPT-3.5 Turbo

GPT-3.5 Turbo is a fast and cost-effective model from OpenAI, optimized for general-purpose conversational tasks and simple reasoning. It supports long context windows and is suitable for a wide range of applications requiring quick responses.

FastCheapLong contextReasoningCoding

Input / 1M tokens

$0.50

Output tokens/s

131.01

First-token seconds

0.82s

Artificial Analysis Intelligence Index

3.6

GPT-3.5 Turbo

GPT-3.5 Turbo (0613)

GPT-3.5 Turbo 0613 is a stable and widely-used version of the GPT-3.5 series, offering a strong balance between performance, speed, and cost. It is well-suited for general-purpose conversational AI, text generation, and analysis tasks, and was the first version to support function calling.

CodingReasoningFastCheap

Input / 1M tokens

$0.00

GPT-4

GPT-4

GPT-4 is a large multimodal model that can process both text and images. It demonstrates strong reasoning capabilities, supports long context windows, and excels at complex tasks including coding and analysis.

MultimodalReasoningCodingLong context

Input / 1M tokens

$30.00

Output tokens/s

33.97

First-token seconds

1.04s

Artificial Analysis Intelligence Index

7

GPT-4

GPT-4 Turbo

GPT-4 Turbo is an enhanced version of GPT-4, featuring a significantly larger 128K context window and improved performance. It supports multimodal inputs (text and images) and offers better cost-efficiency and speed compared to the original GPT-4.

CodingReasoningFastCheapLong contextMultimodal

Input / 1M tokens

$10.00

Output tokens/s

33.24

First-token seconds

1.35s

Artificial Analysis Intelligence Index

7.9

GPT-4.1

GPT-4.1

GPT-4.1 is an advanced multimodal model from OpenAI, likely an iteration or variant within the GPT-4 family. It is designed for strong reasoning and complex task-solving, supporting both text and image inputs.

ReasoningMultimodal

Input / 1M tokens

$2.00

Output tokens/s

121.04

First-token seconds

0.68s

Artificial Analysis Intelligence Index

19.4

GPT-4.1

GPT-4.1 mini

GPT-4.1 mini is a fast and cost-effective model from the GPT-4 family, optimized for everyday tasks and quick responses. It retains strong reasoning and coding capabilities while offering lower latency and pricing.

FastCheapReasoningCoding

Input / 1M tokens

$0.40

Output tokens/s

74.41

First-token seconds

0.62s

Artificial Analysis Intelligence Index

14.8

GPT-4.1

GPT-4.1 nano

GPT-4.1 nano is a lightweight, cost-effective model in the GPT-4 series, optimized for high throughput and low-latency applications. It provides a balance of basic reasoning and coding capabilities at a significantly reduced cost and faster response speed compared to larger models.

FastCheap

Input / 1M tokens

$0.10

Output tokens/s

145.88

First-token seconds

0.54s

Artificial Analysis Intelligence Index

9.6

GPT-4.5

GPT-4.5 (Preview)

This is an upgraded preview model from OpenAI's GPT-4 series, offering significant performance improvements in reasoning, coding, and multimodal tasks. It supports a 128k context window.

ReasoningCodingMultimodalLong context

Input / 1M tokens

$0.00

Artificial Analysis Intelligence Index

13.6

GPT-4o

GPT-4o (Aug '24)

GPT-4o (Aug '24) is OpenAI's flagship multimodal model, capable of processing and generating text, images, and audio. It offers a strong balance of high-level reasoning, fast response times, and a large context window, making it suitable for complex, real-time applications.

MultimodalReasoningFastLong context

Input / 1M tokens

$2.50

Output tokens/s

98.47

First-token seconds

0.63s

Artificial Analysis Intelligence Index

9.6

GPT-4o

GPT-4o (ChatGPT)

GPT-4o is OpenAI's multimodal flagship model, capable of processing text, images, and audio inputs. It offers significantly faster response speeds and lower costs compared to previous models while maintaining strong performance across reasoning, coding, and creative tasks.

CodingReasoningFastCheapMultimodal

Input / 1M tokens

$0.00

Artificial Analysis Intelligence Index

8.2

GPT-4o

GPT-4o (March 2025, chatgpt-4o-latest)

GPT-4o is OpenAI's multimodal flagship model, offering fast response times and strong reasoning capabilities. It supports text, image, and audio inputs, providing a comprehensive and real-time interactive experience. This March 2025 version represents the latest iteration of the GPT-4o series.

CodingReasoningFastLong contextMultimodal

Input / 1M tokens

$0.00

Artificial Analysis Intelligence Index

12.3

GPT-4o

GPT-4o (May '24)

GPT-4o is a multimodal model from OpenAI that can process text, images, and audio. It offers fast response times and improved cost-efficiency compared to previous flagship models.

MultimodalFastReasoningCheap

Input / 1M tokens

$5.00

Output tokens/s

94.8

First-token seconds

0.69s

Artificial Analysis Intelligence Index

8.6

GPT-4o

GPT-4o (Nov '24)

GPT-4o is OpenAI's flagship multimodal model, capable of processing and generating text, images, and audio. It offers significantly faster response speeds and improved reasoning capabilities compared to its predecessors, while maintaining strong performance across coding, analysis, and creative tasks.

CodingReasoningFastLong contextMultimodal

Input / 1M tokens

$2.50

Output tokens/s

199.73

First-token seconds

0.57s

Artificial Analysis Intelligence Index

11.2

GPT-4o

GPT-4o Realtime (Dec '24)

GPT-4o Realtime (Dec '24) is a multimodal model optimized for real-time, low-latency voice and text interactions. It supports simultaneous processing of audio, text, and vision inputs, making it ideal for conversational AI and interactive applications.

MultimodalFastReasoningLong context

Input / 1M tokens

$0.00

GPT-4o

GPT-4o mini

GPT-4o mini is a lightweight, fast, and cost-effective multimodal model from OpenAI. It is designed for high-throughput applications, offering strong performance in reasoning and coding tasks while supporting text and image inputs.

FastCheapMultimodalCodingReasoning

Input / 1M tokens

$0.15

Output tokens/s

73.39

First-token seconds

0.68s

Artificial Analysis Intelligence Index

6.9

GPT-4o

GPT-4o mini Realtime (Dec '24)

A variant of the GPT-4o mini model optimized for low-latency, real-time interactions such as voice conversations and streaming applications. It maintains the core strengths of the GPT-4o mini family, including speed and cost-efficiency, while being specifically tuned for scenarios requiring immediate response and continuous data flow.

CodingReasoningFastCheapMultimodal

Input / 1M tokens

$0.00

GPT-5

GPT-5 (ChatGPT)

GPT-5 is OpenAI's next-generation large language model, expected to feature significant advancements in reasoning, multimodal understanding, and long-context processing. It aims to deliver more accurate, coherent, and versatile responses across complex tasks.

CodingReasoningLong contextMultimodal

Input / 1M tokens

$1.25

Output tokens/s

171.92

First-token seconds

0.57s

Artificial Analysis Intelligence Index

15.3

GPT-5

GPT-5 (high)

OpenAI is a leading AI research and deployment company that develops advanced language models such as GPT-5, GPT-5.4, and GPT-5.5. These models are designed for coding, reasoning, agentic tasks, and support multimodal inputs and long context windows. The company focuses on AI safety and innovation in the AI market.

Input / 1M tokens

$1.25

Output tokens/s

86.61

First-token seconds

79.18s

Artificial Analysis Intelligence Index

34.7

GPT-5

GPT-5 (low)

A lightweight, cost-optimized variant of the GPT-5 model family, designed for high-throughput applications. It retains strong reasoning and coding capabilities while offering faster response times and lower operational costs compared to the full-scale model.

CodingReasoningFastCheapLong contextMultimodal

Input / 1M tokens

$1.25

Output tokens/s

76.21

First-token seconds

8.55s

Artificial Analysis Intelligence Index

31.2

GPT-5

GPT-5 (medium)

GPT-5 (medium) is a mid-tier model in OpenAI's GPT-5 family, offering a strong balance of advanced reasoning, coding proficiency, and multimodal understanding. It supports extended context windows and provides fast response times, making it suitable for complex, multi-step tasks.

ReasoningCodingMultimodalLong contextFast

Input / 1M tokens

$1.25

Output tokens/s

80.71

First-token seconds

50.83s

Artificial Analysis Intelligence Index

33.7

GPT-5

GPT-5 (minimal)

GPT-5 (minimal) is a lightweight, cost-optimized variant of the GPT-5 model family. It is designed for high-speed inference and lower operational costs while retaining the core reasoning and multimodal capabilities of the GPT-5 series.

FastCheapReasoningCodingMultimodal

Input / 1M tokens

$1.25

Output tokens/s

64.07

First-token seconds

0.92s

Artificial Analysis Intelligence Index

17.2

GPT-5

GPT-5 Codex (high)

GPT-5 Codex (high) is a high-performance model from OpenAI's Codex series, optimized for advanced code generation, understanding, and complex reasoning tasks. It is designed to handle sophisticated programming challenges with high accuracy and efficiency.

CodingReasoning

Input / 1M tokens

$1.25

Output tokens/s

167.78

First-token seconds

6.14s

Artificial Analysis Intelligence Index

36.1

GPT-5

GPT-5 mini (high)

GPT-5 mini (high) is a high-performance, compact variant within the GPT-5 family. It is optimized for a balance of strong reasoning and coding capabilities with improved speed and cost-efficiency compared to larger models.

ReasoningCodingFastCheap

Input / 1M tokens

$0.25

Output tokens/s

92.1

First-token seconds

89.5s

Artificial Analysis Intelligence Index

25.3

GPT-5

GPT-5 mini (medium)

A mid-sized model in the GPT-5 family, designed to offer a strong balance between performance, speed, and cost. It is likely optimized for efficient inference while maintaining robust capabilities across general tasks.

FastCheap

Input / 1M tokens

$0.25

Output tokens/s

96.28

First-token seconds

15.21s

Artificial Analysis Intelligence Index

30.9

GPT-5

GPT-5 mini (minimal)

GPT-5 mini (minimal) is a lightweight, cost-effective variant of the GPT-5 family, optimized for fast response times and lower operational costs. It retains the core reasoning and multimodal capabilities of the GPT-5 series while being positioned as an accessible option for high-throughput or budget-conscious applications.

CheapFastReasoningCodingMultimodal

Input / 1M tokens

$0.25

Output tokens/s

101.68

First-token seconds

0.74s

Artificial Analysis Intelligence Index

14.3

GPT-5

GPT-5 nano (high)

GPT-5 nano is the most lightweight and cost-effective model in the GPT-5 family, optimized for speed and low latency while retaining strong reasoning capabilities. It is designed for high-throughput, low-cost applications where rapid response is critical.

FastCheapReasoningCoding

Input / 1M tokens

$0.05

Output tokens/s

149

First-token seconds

98.44s

Artificial Analysis Intelligence Index

19.9

GPT-5

GPT-5 nano (medium)

This is a lightweight, medium-sized model in the GPT-5 series, designed to offer a balance of performance, speed, and cost-effectiveness. It inherits core GPT-5 capabilities such as reasoning and code generation, while being optimized for broader deployment scenarios.

ReasoningFastCheapCoding

Input / 1M tokens

$0.05

Output tokens/s

153.07

First-token seconds

49.57s

Artificial Analysis Intelligence Index

19

GPT-5

GPT-5 nano (minimal)

GPT-5 nano (minimal) is a lightweight variant of the GPT-5 series, designed for fast inference and low operational cost. It maintains essential reasoning and coding abilities while being optimized for edge or resource-limited deployments.

FastCheapReasoning

Input / 1M tokens

$0.05

Output tokens/s

148.41

First-token seconds

0.68s

Artificial Analysis Intelligence Index

8

GPT-5.1

GPT-5.1 (Non-reasoning)

This is a non-reasoning variant of the GPT-5.1 series, optimized for speed and cost. It retains multimodal, long-context, and coding capabilities but provides faster responses and lower usage costs by reducing reasoning steps.

CodingFastCheapLong contextMultimodal

Input / 1M tokens

$1.25

Output tokens/s

92.57

First-token seconds

0.69s

Artificial Analysis Intelligence Index

20.4

GPT-5.1

GPT-5.1 (high)

GPT-5.1 (high) is a high-performance variant of OpenAI's GPT-5 series, optimized for advanced reasoning and complex problem-solving tasks. It likely represents a more capable or resource-intensive configuration within the GPT-5 family.

CodingReasoning

Input / 1M tokens

$1.25

Output tokens/s

94.05

First-token seconds

24.83s

Artificial Analysis Intelligence Index

36.9

GPT-5.1

GPT-5.1 Codex (high)

GPT-5.1 Codex (high) is a specialized model from OpenAI optimized for advanced code generation, understanding, and manipulation tasks. It is designed to handle complex programming challenges with high accuracy and efficiency, likely featuring strong reasoning capabilities for algorithmic and logical problems.

CodingReasoning

Input / 1M tokens

$1.25

Output tokens/s

166.48

First-token seconds

3.4s

Artificial Analysis Intelligence Index

34.7

GPT-5.1

GPT-5.1 Codex mini (high)

A high-performance, lightweight model from the Codex series optimized for code generation and understanding. It offers a strong balance of speed, cost-efficiency, and coding capability, making it suitable for fast, high-volume code-related tasks.

CodingFastCheapLong context

Input / 1M tokens

$0.25

Output tokens/s

213.48

First-token seconds

3.73s

Artificial Analysis Intelligence Index

30.6

GPT-5.2

GPT-5.2 (Non-reasoning)

A high-performance variant of the GPT-5 series optimized for speed and cost-efficiency. It is designed for general-purpose tasks, offering strong multimodal and coding capabilities without the overhead of extended reasoning chains.

CodingFastCheapLong contextMultimodal

Input / 1M tokens

$1.75

Output tokens/s

66.79

First-token seconds

0.75s

Artificial Analysis Intelligence Index

26

GPT-5.2

GPT-5.2 (medium)

GPT-5.2 (medium) is a balanced model within the GPT-5 series, offering strong reasoning and coding capabilities with multimodal support. It is designed to provide a good trade-off between performance, speed, and cost for a wide range of general-purpose tasks.

ReasoningFastMultimodalCoding

Input / 1M tokens

$1.75

Artificial Analysis Intelligence Index

38

GPT-5.2

GPT-5.2 (xhigh)

GPT-5.2 (xhigh) is a high-performance variant of OpenAI's GPT-5 series, offering enhanced reasoning, multimodal understanding, and long-context capabilities. It is designed for complex, high-stakes tasks requiring deep analysis and synthesis across text and images.

ReasoningCodingMultimodalLong context

Input / 1M tokens

$1.75

Output tokens/s

73.2

First-token seconds

71.4s

Artificial Analysis Intelligence Index

42.2

GPT-5.2

GPT-5.2 Codex (xhigh)

GPT-5.2 Codex (xhigh) is a high-performance model from OpenAI's Codex series, specialized in advanced code generation, understanding, and transformation tasks. It is optimized for complex programming challenges and likely features enhanced reasoning capabilities for technical domains.

CodingReasoningFast

Input / 1M tokens

$1.75

Output tokens/s

121.97

First-token seconds

1.17s

Artificial Analysis Intelligence Index

40.1

GPT-5.3 Codex

GPT-5.3 Codex (xhigh)

A high-performance code generation model from OpenAI's GPT-5 series, specifically optimized for coding tasks with fast response times and strong reasoning capabilities.

CodingReasoningFastLong context

Input / 1M tokens

$1.75

Output tokens/s

77.07

First-token seconds

84.45s

Artificial Analysis Intelligence Index

44.3

GPT-5.4

GPT-5.4 (Non-reasoning)

GPT-5.4 (Non-reasoning) is a high-performance, general-purpose model optimized for speed and cost-efficiency. It delivers strong core capabilities in coding, multimodal understanding, and long-context processing without the overhead of advanced reasoning or chain-of-thought features, making it ideal for high-throughput applications.

CodingFastCheapLong contextMultimodal

Input / 1M tokens

$2.50

Output tokens/s

104.32

First-token seconds

0.69s

Artificial Analysis Intelligence Index

27.7

GPT-5.4

GPT-5.4 (low)

OpenAI is a leading AI research organization that develops advanced language models like GPT-5.4, which is designed for complex professional work including coding, reasoning, and long-context tasks. The company focuses on creating powerful AI systems with capabilities in multimodal and fast inference applications.

Input / 1M tokens

$2.50

Output tokens/s

106.13

First-token seconds

1.15s

Artificial Analysis Intelligence Index

39.1

GPT-5.4

GPT-5.4 (xhigh)

GPT-5.4 (xhigh) is a high-performance variant within OpenAI's GPT-5 series, designed for advanced reasoning, complex coding tasks, and multimodal understanding. It likely supports extended context windows and excels in generating nuanced, high-quality outputs.

CodingReasoningLong contextMultimodal

Input / 1M tokens

$2.50

Output tokens/s

156.64

First-token seconds

119.14s

Artificial Analysis Intelligence Index

51.4

GPT-5.4

GPT-5.4 Pro (xhigh)

GPT-5.4 Pro (xhigh) is a high-performance variant of OpenAI's GPT-5 series, optimized for advanced reasoning, complex problem-solving, and extended context handling. It features enhanced multimodal capabilities and is designed for demanding enterprise and research applications.

CodingReasoningFastLong contextMultimodal

Input / 1M tokens

$30.00

GPT-5.4

GPT-5.4 mini (Non-Reasoning)

A smaller, non-reasoning variant of the GPT-5.4 model from OpenAI. It is optimized for fast, low-cost responses in tasks that do not require complex chain-of-thought or deep reasoning capabilities.

CodingFastCheap

Input / 1M tokens

$0.75

Output tokens/s

158.22

First-token seconds

0.57s

Artificial Analysis Intelligence Index

16.6

GPT-5.4

GPT-5.4 mini (medium)

GPT-5.4 mini (medium) is a balanced, lightweight variant within the GPT-5 series, optimized for a strong combination of performance and cost-efficiency. It retains core reasoning and multimodal capabilities while offering faster response times and lower operational costs compared to larger models in the family.

CodingReasoningFastCheapMultimodal

Input / 1M tokens

$0.75

Output tokens/s

165.6

First-token seconds

6.44s

Artificial Analysis Intelligence Index

29.8

GPT-5.4

GPT-5.4 mini (xhigh)

OpenAI is an AI research and deployment company known for developing large language models like the GPT series. Their products include GPT-5.4 and optimized variants such as GPT-5.4 mini for high-volume, low-latency applications in coding, tool use, and multimodal reasoning.

Input / 1M tokens

$0.75

Output tokens/s

153.38

First-token seconds

6.99s

Artificial Analysis Intelligence Index

40

GPT-5.4

GPT-5.4 nano (Non-Reasoning)

A lightweight, non-reasoning variant of the GPT-5 series optimized for speed and cost-efficiency. Designed for straightforward tasks where rapid response and low latency are prioritized over complex chain-of-thought reasoning.

FastCheap

Input / 1M tokens

$0.20

Output tokens/s

168.09

First-token seconds

0.56s

Artificial Analysis Intelligence Index

17.6

GPT-5.4

GPT-5.4 nano (medium)

A lightweight and efficient model from the GPT-5.4 series, optimized for fast response times and low operational costs. It is designed for high-throughput tasks where speed and affordability are primary considerations.

FastCheapReasoning

Input / 1M tokens

$0.20

Output tokens/s

166.29

First-token seconds

3.09s

Artificial Analysis Intelligence Index

30.2

GPT-5.4

GPT-5.4 nano (xhigh)

OpenAI is a leading AI research and deployment company that develops advanced language models like the GPT-5.4 series, including the nano variant optimized for high-speed, cost-effective tasks such as classification and sub-agent workloads.

Input / 1M tokens

$0.20

Output tokens/s

158.53

First-token seconds

2.47s

Artificial Analysis Intelligence Index

38.2

GPT-5.5

GPT-5.5 (Non-reasoning)

A variant of the GPT-5.5 model optimized for fast, direct responses without extended reasoning or chain-of-thought processes. It is designed for low-latency applications where speed and cost-efficiency are prioritized over complex, step-by-step problem solving.

FastCheap

Input / 1M tokens

$5.00

Output tokens/s

61.1

First-token seconds

0.88s

Artificial Analysis Intelligence Index

35.4

GPT-5.5

GPT-5.5 (high)

GPT-5.5 (high) is a high-performance variant of OpenAI's next-generation GPT-5 model family. It is designed to deliver advanced reasoning, coding, and multimodal capabilities, likely representing a top-tier configuration within its series.

CodingReasoningMultimodal

Input / 1M tokens

$5.00

Output tokens/s

63.51

First-token seconds

20.65s

Artificial Analysis Intelligence Index

53.1

GPT-5.5

GPT-5.5 (low)

GPT-5.5 (low) is a lightweight, cost-optimized variant within the GPT-5.5 series. It is designed for high-throughput applications, emphasizing fast response times and low operational costs while retaining strong reasoning and coding capabilities.

CodingReasoningFastCheap

Input / 1M tokens

$5.00

Output tokens/s

60.69

First-token seconds

1.75s

Artificial Analysis Intelligence Index

43.5

GPT-5.5

GPT-5.5 (medium)

OpenAI is an AI research and deployment company that develops advanced language models like GPT-5.5. Their models are designed for complex tasks such as coding, reasoning, and data analysis, and are accessible via API.

Input / 1M tokens

$5.00

Output tokens/s

61.02

First-token seconds

4.57s

Artificial Analysis Intelligence Index

50.4

GPT-5.5

GPT-5.5 (xhigh)

GPT-5.5 (xhigh) is a high-performance variant of OpenAI's GPT-5 series, optimized for superior reasoning, complex code generation, and advanced multimodal tasks. It represents a flagship model designed for demanding applications requiring top-tier intelligence and capability.

CodingReasoningMultimodal

Input / 1M tokens

$5.00

Output tokens/s

63.03

First-token seconds

35.79s

Artificial Analysis Intelligence Index

54.8

OpenAI

GPT-5.5 Instant (June 2026)

Input / 1M tokens

$5.00

Artificial Analysis Intelligence Index

28.9

OpenAI

GPT-5.5 Instant (May 2026)

OpenAI is a leading AI research and deployment company that develops advanced AI models. Its GPT-5.5 Instant model is the current default for ChatGPT, offering improved capabilities in coding, reasoning, and personalization with reduced hallucinations. The company provides these models through API services for various applications.

Input / 1M tokens

$5.00

Artificial Analysis Intelligence Index

33.5

GPT-5.5

GPT-5.5 Pro (xhigh)

OpenAI develops advanced AI models like GPT-5.5 for complex professional and research tasks, offering API access with features such as variable reasoning effort and multimodal capabilities.

Input / 1M tokens

$0.00

OpenAI

GPT-5.6 Luna (Non-reasoning)

Input / 1M tokens

$1.00

Output tokens/s

207.01

First-token seconds

0.58s

Artificial Analysis Intelligence Index

26.6

OpenAI

GPT-5.6 Luna (high)

Input / 1M tokens

$1.00

Output tokens/s

204.91

First-token seconds

5.28s

Artificial Analysis Intelligence Index

46.1

OpenAI

GPT-5.6 Luna (low)

Input / 1M tokens

$1.00

Output tokens/s

232.91

First-token seconds

1.03s

Artificial Analysis Intelligence Index

33.3

OpenAI

GPT-5.6 Luna (max)

Input / 1M tokens

$1.00

Output tokens/s

231.35

First-token seconds

69.12s

Artificial Analysis Intelligence Index

51.2

OpenAI

GPT-5.6 Luna (medium)

Input / 1M tokens

$1.00

Output tokens/s

231.72

First-token seconds

1.61s

Artificial Analysis Intelligence Index

38.1

OpenAI

GPT-5.6 Luna (xhigh)

Input / 1M tokens

$1.00

Output tokens/s

208.16

First-token seconds

25.11s

Artificial Analysis Intelligence Index

49.1

OpenAI

GPT-5.6 Sol (Non-reasoning)

Input / 1M tokens

$5.00

Artificial Analysis Intelligence Index

41.2

OpenAI

GPT-5.6 Sol (high)

Input / 1M tokens

$5.00

Output tokens/s

74.43

First-token seconds

10.26s

Artificial Analysis Intelligence Index

55.9

OpenAI

GPT-5.6 Sol (low)

Input / 1M tokens

$5.00

Output tokens/s

87.55

First-token seconds

1.93s

Artificial Analysis Intelligence Index

49.4

OpenAI

GPT-5.6 Sol (max)

Input / 1M tokens

$5.00

Output tokens/s

87.48

First-token seconds

76.93s

Artificial Analysis Intelligence Index

58.9

OpenAI

GPT-5.6 Sol (medium)

Input / 1M tokens

$5.00

Output tokens/s

65.49

First-token seconds

4.52s

Artificial Analysis Intelligence Index

53.6

OpenAI

GPT-5.6 Sol (xhigh)

Input / 1M tokens

$5.00

Output tokens/s

87.77

First-token seconds

7.6s

Artificial Analysis Intelligence Index

57.7

OpenAI

GPT-5.6 Terra (Non-reasoning)

Input / 1M tokens

$2.50

Output tokens/s

132.36

First-token seconds

0.79s

Artificial Analysis Intelligence Index

34

OpenAI

GPT-5.6 Terra (high)

Input / 1M tokens

$2.50

Output tokens/s

124.51

First-token seconds

1.82s

Artificial Analysis Intelligence Index

49

OpenAI

GPT-5.6 Terra (low)

Input / 1M tokens

$2.50

Output tokens/s

132.18

First-token seconds

1.13s

Artificial Analysis Intelligence Index

40.5

OpenAI

GPT-5.6 Terra (max)

Input / 1M tokens

$2.50

Output tokens/s

149.7

First-token seconds

109.95s

Artificial Analysis Intelligence Index

55

OpenAI

GPT-5.6 Terra (medium)

Input / 1M tokens

$2.50

Output tokens/s

133.56

First-token seconds

1.05s

Artificial Analysis Intelligence Index

45.6

OpenAI

GPT-5.6 Terra (xhigh)

Input / 1M tokens

$2.50

Output tokens/s

129.36

First-token seconds

4.86s

Artificial Analysis Intelligence Index

51.6

OpenAI

gpt-oss-120b (high)

OpenAI is a leading AI research and deployment company. It develops advanced AI models such as the open-weight gpt-oss-120b, which uses a mixture-of-experts architecture for efficient high-reasoning capabilities. The company offers AI products via its API and platform for various applications.

Input / 1M tokens

$0.15

Output tokens/s

248.8

First-token seconds

0.52s

Artificial Analysis Intelligence Index

23.8

OpenAI

gpt-oss-120b (low)

GPT-OSS-120B (low) is a large language model from OpenAI with 120 billion parameters, optimized for lower resource usage while maintaining strong reasoning and coding capabilities.

ReasoningCoding

Input / 1M tokens

$0.15

Output tokens/s

350.98

First-token seconds

0.52s

Artificial Analysis Intelligence Index

17.7

OpenAI

gpt-oss-20b (high)

A 20-billion parameter open-source language model from OpenAI, optimized for high performance. It is designed for strong coding and reasoning capabilities while maintaining fast inference speeds.

CodingReasoningFastCheap

Input / 1M tokens

$0.05

Output tokens/s

202.14

First-token seconds

0.42s

Artificial Analysis Intelligence Index

14.9

OpenAI

gpt-oss-20b (low)

Based on its naming, this appears to be a 20-billion parameter open-source model variant optimized for low-resource or cost-efficient deployment. It is likely part of a community or third-party series inspired by GPT architectures, rather than an official OpenAI release.

FastCheap

Input / 1M tokens

$0.06

Output tokens/s

236.93

First-token seconds

0.46s

Artificial Analysis Intelligence Index

14.3

o1

o1

OpenAI o1 is a reasoning model designed to solve complex problems in science, coding, and mathematics. It employs an internal chain-of-thought process to analyze questions before responding, significantly improving accuracy on tasks requiring deep logical inference.

ReasoningCodingLong context

Input / 1M tokens

$15.00

Output tokens/s

141.32

First-token seconds

14.51s

Artificial Analysis Intelligence Index

23.4

o1

o1-mini

A smaller, faster, and more cost-effective reasoning model from OpenAI's o1 family. It excels at complex tasks in math, coding, and science while maintaining strong performance with reduced latency and cost.

CodingReasoningFastCheapLong context

Input / 1M tokens

$0.00

Artificial Analysis Intelligence Index

14

o1

o1-preview

o1-preview is OpenAI's first reasoning model, designed to solve complex problems in science, coding, and mathematics. It uses an internal chain-of-thought process to analyze and reason through multi-step tasks before responding, significantly improving accuracy on challenging queries.

ReasoningCoding

Input / 1M tokens

$16.50

Artificial Analysis Intelligence Index

17

o1

o1-pro

OpenAI o1-pro is a professional-grade reasoning model from the o1 family, designed for complex problem-solving tasks. It features an advanced chain-of-thought process to tackle challenging questions in science, math, and coding with high accuracy.

ReasoningCoding

Input / 1M tokens

$150.00

Artificial Analysis Intelligence Index

18.9

o3

o3

OpenAI o3 is the latest model in the o-series, designed for advanced reasoning tasks. It excels at solving complex problems in mathematics, science, and coding by utilizing an extended chain-of-thought process. As a successor to o1, it represents a significant step forward in AI's ability to perform deep, deliberate thinking.

ReasoningCoding

Input / 1M tokens

$2.00

Output tokens/s

107.11

First-token seconds

6.81s

Artificial Analysis Intelligence Index

30.4

o3

o3-mini

o3-mini is a lightweight reasoning model from OpenAI's o3 family, optimized for fast and cost-effective performance on complex tasks like coding and math. It retains strong reasoning capabilities while offering lower latency and reduced costs compared to larger models.

CodingReasoningFastCheap

Input / 1M tokens

$1.10

Output tokens/s

226.88

First-token seconds

5.04s

Artificial Analysis Intelligence Index

19

o3

o3-mini (high)

A high-performance variant of OpenAI's o3-mini model, optimized for complex reasoning and coding tasks. It offers a strong balance of intelligence, speed, and cost-efficiency within the o3 family, supporting long context windows.

ReasoningCodingFastCheapLong context

Input / 1M tokens

$1.10

Output tokens/s

217.94

First-token seconds

16.24s

Artificial Analysis Intelligence Index

15.6

o3

o3-pro

An enhanced reasoning model from OpenAI's o-series, designed for complex problem-solving with deep chain-of-thought capabilities. It excels in tasks requiring multi-step logical inference and analysis.

ReasoningCoding

Input / 1M tokens

$20.00

Output tokens/s

28.47

First-token seconds

63.54s

Artificial Analysis Intelligence Index

32.5

o4-mini

o4-mini (high)

This is a lightweight reasoning model from OpenAI's o4 series, designed for efficient handling of complex reasoning tasks. It leverages an internal thinking mechanism to enhance performance in mathematics, coding, and scientific problems.

ReasoningFastCheapCoding

Input / 1M tokens

$1.10

Output tokens/s

157.08

First-token seconds

18.39s

Artificial Analysis Intelligence Index

25.6

Discussion

Thinking... Make sure you are connected to GitHub server