# OpenCode Data: Real-world AI model usage, cache ratios, and costs

> Source: <https://opencode.ai/data/>
> Published: 2026-07-08 14:32:00+00:00

[Model Data](#overview)

See which models are winning real usage, how the mix is shifting, and what that means for cost.

Top Models. Usage of models across OpenCode Go.

[Top Models.](#top-models)

[04](/data/minimax/minimax-m3)

**minimax-m3** 960B

MiniMax-61%

[05](/data/xiaomi/mimo-v2-5)

**mimo-v2.5** 827B

Xiaomi-4%

[06](/data/moonshot/kimi-k2-7-code)

**kimi-k2.7-code** 391B

Moonshot-17%

[07](/data/xiaomi/mimo-v2-5-pro)

**mimo-v2.5-pro** 273B

Xiaomi-5%

[08](/data/qwen/qwen3-7-plus)

**qwen3.7-plus** 247B

Qwen+15%

[09](/data/moonshot/kimi-k2-6)

**kimi-k2.6** 118B

Moonshot-15%

[10](/data/zhipu/glm-5-1)

**glm-5.1** 59B

Zhipu-20%

[11](/data/minimax/minimax-m2-7)

**minimax-m2.7** 54B

MiniMax-14%

[12](/data/qwen/qwen3-7-max)

**qwen3.7-max** 54B

Qwen-4%

[13](/data/qwen/qwen3-6-plus)

**qwen3.6-plus** 39B

Qwen-2%

[14](/data/moonshot/kimi-k2-5)

**kimi-k2.5** 6B

Moonshot-19%

[15](/data/zhipu/glm-5)

**glm-5** 5B

Zhipu-14%

[16](/data/minimax/minimax-m2-5)

**minimax-m2.5** 5B

MiniMax-7%

[17](/data/qwen/qwen3-5-plus)

**qwen3.5-plus** 2B

Qwen-45%

[18](/data/tencent/hy3-preview)

**hy3-preview** 0B

Tencent-58%

[01](/data/deepseek/deepseek-v4-flash)

**deepseek-v4-flash** 8.2T

DeepSeek-20%

[02](/data/deepseek/deepseek-v4-pro)

**deepseek-v4-pro** 2.8T

DeepSeek-27%

[03](/data/zhipu/glm-5-2)

**glm-5.2** 1.3T

Zhipu-17%

[04](/data/minimax/minimax-m3)

**minimax-m3** 960B

MiniMax-61%

[05](/data/xiaomi/mimo-v2-5)

**mimo-v2.5** 827B

Xiaomi-4%

[06](/data/moonshot/kimi-k2-7-code)

**kimi-k2.7-code** 391B

Moonshot-17%

[07](/data/xiaomi/mimo-v2-5-pro)

**mimo-v2.5-pro** 273B

Xiaomi-5%

[08](/data/qwen/qwen3-7-plus)

**qwen3.7-plus** 247B

Qwen+15%

[09](/data/moonshot/kimi-k2-6)

**kimi-k2.6** 118B

Moonshot-15%

[10](/data/zhipu/glm-5-1)

**glm-5.1** 59B

Zhipu-20%

[11](/data/minimax/minimax-m2-7)

**minimax-m2.7** 54B

MiniMax-14%

[12](/data/qwen/qwen3-7-max)

**qwen3.7-max** 54B

Qwen-4%

[13](/data/qwen/qwen3-6-plus)

**qwen3.6-plus** 39B

Qwen-2%

[14](/data/moonshot/kimi-k2-5)

**kimi-k2.5** 6B

Moonshot-19%

[15](/data/zhipu/glm-5)

**glm-5** 5B

Zhipu-14%

[16](/data/minimax/minimax-m2-5)

**minimax-m2.5** 5B

MiniMax-7%

[17](/data/qwen/qwen3-5-plus)

**qwen3.5-plus** 2B

Qwen-45%

[18](/data/tencent/hy3-preview)

**hy3-preview** 0B

Tencent-58%

[LEARN MORE](#top-models)

**TOP MODELS****▸**** Unique Users.** Daily unique OpenCode Go users by model.