AI models

16 models · 6 providers · One endpoint

Every model below is callable through the same OpenAI-compatible API and the same key. Prices are indicative — USD per 1M tokens for text models, USD per image for image models. — 1 of them free

Provider

Capability

Showing 16 of 16 models

OpenAI

GPT-5.2

Featured

ChatGPT’s flagship — strong general reasoning, vision and tool use.

ChatReasoningTool CallingVision

Context: 391K
Input / 1M: $0.88$1.75 list
Output / 1M: $7$14 list

gpt-5.2Try it

OpenAI

GPT-5.2 mini

Fast, cheap ChatGPT tier for high-volume chat, classification and extraction.

ChatTool Calling

Context: 391K
Input / 1M: $0.13$0.25 list
Output / 1M: $1$2 list

gpt-5.2-miniTry it

DeepSeek

DeepSeek-V3.2

Featured

Flagship non-thinking model. Fast, cheap, and excellent at general chat and tool use.

ChatTool CallingLong Context

Context: 128K
Input / 1M: $0.13$0.27 list
Output / 1M: $0.20$0.40 list

deepseek-chatTry it

DeepSeek

DeepSeek-R1

Reasoning model with visible chain-of-thought for math, code, and hard problems.

ReasoningTool CallingLong Context

Context: 128K
Input / 1M: $0.35$0.70 list
Output / 1M: $1.25$2.50 list

deepseek-reasonerTry it

Zhipu GLM

GLM-4.6

Featured

Z.ai flagship. Strong coding, writing, and agentic planning; 200K context.

ChatReasoningTool CallingLong Context

Context: 200K
Input / 1M: $0.30$0.60 list
Output / 1M: $1.10$2.20 list

glm-4.6Try it

Zhipu GLM

GLM-4.5 Air

Lighter GLM tuned for cost-sensitive production traffic.

ChatTool Calling

Context: 128K
Input / 1M: $0.10$0.20 list
Output / 1M: $0.55$1.10 list

glm-4.5-airTry it

Zhipu GLM

GLM-4.5 Flash

Free tier of the GLM family — great for prototypes and hobby projects.

ChatTool Calling

Context: 128K
Input / 1M: FreeFree list
Output / 1M: FreeFree list

glm-4.5-flashTry it

Alibaba Qwen

Qwen3-Max

Alibaba’s largest flagship. Top-tier quality for demanding agentic and chat workloads.

ChatTool CallingLong Context

Context: 256K
Input / 1M: $0.60$1.20 list
Output / 1M: $3$6 list

qwen3-maxTry it

Alibaba Qwen

Qwen3 235B A22B

Open-weights MoE flagship with hybrid thinking mode.

ChatReasoningTool CallingLong Context

Context: 256K
Input / 1M: $0.10$0.20 list
Output / 1M: $0.40$0.80 list

qwen3-235b-a22bTry it

Alibaba Qwen

Qwen3 32B

Dense open-weights all-rounder with thinking and non-thinking modes.

ChatReasoningTool Calling

Context: 128K
Input / 1M: $0.04$0.08 list
Output / 1M: $0.14$0.28 list

qwen3-32bTry it

Alibaba Qwen

Qwen3 Coder Plus

Code-specialized model built for agentic coding tools and repository-scale edits.

ChatTool CallingLong Context

Context: 256K
Input / 1M: $0.33$0.65 list
Output / 1M: $1.63$3.25 list

qwen3-coder-plusTry it

Alibaba Qwen

Qwen VL Max

Vision-language model for image understanding, OCR, and chart reasoning.

ChatVision

Context: 128K
Input / 1M: $0.14$0.28 list
Output / 1M: $0.42$0.83 list

qwen-vl-maxTry it

Moonshot Kimi

Kimi K2 (0905)

Trillion-parameter MoE tuned for agentic tool use and long documents.

ChatTool CallingLong Context

Context: 256K
Input / 1M: $0.30$0.60 list
Output / 1M: $1.25$2.50 list

kimi-k2-0905Try it

Moonshot Kimi

Kimi K2 Thinking

Reasoning variant of K2 with interleaved thinking for multi-step tasks.

ReasoningTool CallingLong Context

Context: 256K
Input / 1M: $0.30$0.60 list
Output / 1M: $1.25$2.50 list

kimi-k2-thinkingTry it

MiniMax

MiniMax-M2

Efficient agentic MoE — a favorite for coding agents on a budget.

ChatTool CallingLong Context

Context: 200K
Input / 1M: $0.15$0.30 list
Output / 1M: $0.60$1.20 list

MiniMax-M2Try it

OpenAI

GPT-Image-2

Featured

OpenAI’s image generation model — prompt-to-image with strong instruction following.

Text-to-Image

Type: OpenAI
Per image: $0.06$0.11 list
Output: Up to 2K

gpt-image-2Try it

Prices are indicative list prices in USD — per 1M tokens for text models, per generated image for image models — and may differ from the rates configured in your console. Context windows are the maximum supported by each upstream model. See the Docs for integration details.