Find the right LLM for your coding workflow

1.

deepseek-v4-pro

DeepSeek

3,734,590,555 tokens in the last 7 days

deepseek-v4-pro supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

2.

deepseek-v4-flash

DeepSeek

2,564,828,785 tokens in the last 7 days

deepseek-v4-flash supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

3.

gpt-5.6-terra

OpenAI

598,613,955 tokens in the last 7 days

gpt-5.6-terra balances intelligence, speed, and cost for everyday agents and production applications.

4.

claude-opus-4-8

Anthropic

291,091,048 tokens in the last 7 days

claude-opus-4-8 is aimed at complex agentic coding, multi-step engineering, and professional knowledge work.

5.

claude-haiku-4-5-20251001

Anthropic

269,998,175 tokens in the last 7 days

claude-haiku-4-5-20251001 prioritizes low latency and cost efficiency for responsive, high-volume workloads.

6.

kimi-k3

Moonshot AI

181,100,789 tokens in the last 7 days

kimi-k3 focuses on long-context coding, tool use, and agents that carry complex tasks across many steps.

7.

claude-sonnet-4-6

Anthropic

138,289,782 tokens in the last 7 days

claude-sonnet-4-6 balances intelligence, speed, and cost for everyday agents and production applications.

8.

claude-fable-5

Anthropic

claude-fable-5 is built for demanding reasoning, deep analysis, and long-horizon agentic work.

9.

claude-opus-4-6

Anthropic

claude-opus-4-6 is aimed at complex agentic coding, multi-step engineering, and professional knowledge work.

10.

claude-opus-4-7

Anthropic

claude-opus-4-7 is aimed at complex agentic coding, multi-step engineering, and professional knowledge work.

11.

claude-opus-5

Anthropic

claude-opus-5 is aimed at complex agentic coding, multi-step engineering, and professional knowledge work.

12.

claude-sonnet-5

Anthropic

claude-sonnet-5 balances intelligence, speed, and cost for everyday agents and production applications.

13.

deepseek-v3

DeepSeek

deepseek-v3 supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

14.

deepseek-v3.1

DeepSeek

deepseek-v3.1 supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

15.

deepseek-v3.2

DeepSeek

deepseek-v3.2 supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

16.

gemini-3.1-flash-lite

Google

gemini-3.1-flash-lite prioritizes low latency and cost efficiency for responsive, high-volume workloads.

17.

gemini-3.1-pro-preview

Google

gemini-3.1-pro-preview combines reasoning and coding with multimodal understanding for text, images, documents, and visual tasks.

18.

gemini-3.5-flash

Google

gemini-3.5-flash balances intelligence, speed, and cost for everyday agents and production applications.

19.

gemini-3.5-flash-lite

Google

gemini-3.5-flash-lite prioritizes low latency and cost efficiency for responsive, high-volume workloads.

20.

gemini-3.6-flash

Google

gemini-3.6-flash balances intelligence, speed, and cost for everyday agents and production applications.

21.

gemini-3.7-flash

Google

gemini-3.7-flash balances intelligence, speed, and cost for everyday agents and production applications.

22.

glm-5

Z.ai

glm-5 focuses on long-context coding, tool use, and agents that carry complex tasks across many steps.

23.

glm-5-turbo

Z.ai

glm-5-turbo focuses on long-context coding, tool use, and agents that carry complex tasks across many steps.

24.

glm-5.1

Z.ai

glm-5.1 focuses on long-context coding, tool use, and agents that carry complex tasks across many steps.

25.

glm-5.2

Z.ai

glm-5.2 focuses on long-context coding, tool use, and agents that carry complex tasks across many steps.

26.

glm-5.3

Z.ai

glm-5.3 focuses on long-context coding, tool use, and agents that carry complex tasks across many steps.

27.

gpt-5.5

OpenAI

gpt-5.5 is designed for complex professional work, coding, research, and tool-driven workflows.

28.

gpt-5.6-luna

OpenAI

gpt-5.6-luna prioritizes low latency and cost efficiency for responsive, high-volume workloads.

29.

gpt-5.6-sol

OpenAI

gpt-5.6-sol is designed for complex professional work, coding, research, and tool-driven workflows.

30.

kimi-k2.6

Moonshot AI

kimi-k2.6 focuses on long-context coding, tool use, and agents that carry complex tasks across many steps.

31.

macaron-v1-coding-venti

Macaron

macaron-v1-coding-venti is listed by Macaron for Coding Models. Review its live capabilities, context window, availability, and API price before integrating it.

32.

macaron-v1-tall

Macaron

macaron-v1-tall is listed by Macaron for Coding Models. Review its live capabilities, context window, availability, and API price before integrating it.

33.

macaron-v1-venti

Macaron

macaron-v1-venti is listed by Macaron for Coding Models. Review its live capabilities, context window, availability, and API price before integrating it.

34.

minimax-m2.5

MiniMax

minimax-m2.5 is aimed at complex agentic coding, multi-step engineering, and professional knowledge work.

35.

minimax-m2.7

MiniMax

minimax-m2.7 is aimed at complex agentic coding, multi-step engineering, and professional knowledge work.

36.

qwen3.5-27b

Alibaba

qwen3.5-27b supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

37.

qwen3.5-35b-a3b

Alibaba

qwen3.5-35b-a3b supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

38.

qwen3.5-397b-a17b

Alibaba

qwen3.5-397b-a17b supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

39.

qwen3.5-flash

Alibaba (China)

qwen3.5-flash supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

40.

qwen3.5-plus

Alibaba

qwen3.5-plus supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

41.

qwen3.6-plus

Alibaba

qwen3.6-plus supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

42.

qwen3.7-max

Alibaba

qwen3.7-max supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

43.

qwen3.7-plus

Alibaba

qwen3.7-plus supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.

44.

qwen3.8-max

Alibaba

qwen3.8-max supports both deeper reasoning and fast responses for coding, multilingual, and agentic workflows.