Model catalog

35 models. 13 providers. One key.

Every model below is addressable through the same endpoint, on your own provider keys, at the provider's exact price. TokenRouter adds routing, never a markup.

Models

Providers

Widest context

1.05M tokens

Cheapest input

$0.02 / 1M

Can't decide? Pass model="auto" and the router picks from the providers you have keys for.

auto: balanced default
auto:cost: cheapest capable model
auto:latency: fastest first token
auto:quality: strongest available

Model

Context

Input / 1M

Output / 1M

Capabilities

OpenAI 8 models

GPT-5

openai/gpt-5

400K

$1.25

cached $0.125

$10

Reasoning · Vision

GPT-5 mini

openai/gpt-5-mini

400K

$0.25

cached $0.025

Reasoning · Vision

GPT-5 nano

openai/gpt-5-nano

400K

$0.05

cached $0.005

$0.4

Reasoning · Vision

GPT-4.1

openai/gpt-4.1

1.05M

cached $0.5

Vision

GPT-4.1 mini

openai/gpt-4.1-mini

1.05M

$0.4

cached $0.1

$1.6

Vision

GPT-4o

openai/gpt-4o

128K

$2.5

cached $1.25

$10

Vision

OpenAI o3

openai/o3

200K

cached $0.5

Reasoning · Vision

Embedding 3 small

openai/text-embedding-3-small

$0.02

—

Embeddings

Anthropic 3 models

Claude Opus 4.1

anthropic/claude-opus-4-1

200K

$15

cached $1.5

$75

Reasoning · Vision

Claude Sonnet 4.5

anthropic/claude-sonnet-4-5

200K

cached $0.3

$15

Reasoning · Vision

Claude Haiku 4.5

anthropic/claude-haiku-4-5

200K

cached $0.1

Reasoning · Vision

Google 3 models

Gemini 2.5 Pro

google/gemini-2.5-pro

1.05M

$1.25

cached $0.31

$10

Reasoning · Vision

Gemini 2.5 Flash

google/gemini-2.5-flash

1.05M

$0.3

cached $0.075

$2.5

Reasoning · Vision

Gemini 2.5 Flash-Lite

google/gemini-2.5-flash-lite

1.05M

$0.1

cached $0.025

$0.4

Vision

xAI 2 models

Grok 4

xai/grok-4

256K

cached $0.75

$15

Reasoning · Vision

Grok 3 mini

xai/grok-3-mini

131K

$0.3

$0.5

Reasoning

DeepSeek 2 models

DeepSeek V3.1

deepseek/deepseek-chat

128K

$0.27

cached $0.07

$1.1

—

DeepSeek R1

deepseek/deepseek-reasoner

128K

$0.55

cached $0.14

$2.19

Reasoning

Mistral 3 models

Mistral Large

mistral/mistral-large-latest

128K

—

Mistral Small

mistral/mistral-small-latest

128K

$0.1

$0.3

—

Codestral

mistral/codestral-latest

256K

$0.3

$0.9

—

Alibaba Qwen 3 models

Qwen Max

qwen/qwen-max

131K

$1.6

$6.4

—

Qwen Plus

qwen/qwen-plus

131K

$0.4

$1.2

Reasoning

Qwen Turbo

qwen/qwen-turbo

1.01M

$0.05

$0.2

—

Zhipu GLM 2 models

GLM-4.5

glm/glm-4.5

128K

$0.6

$2.2

Reasoning

GLM-4.5 Air

glm/glm-4.5-air

128K

$0.2

$1.1

Reasoning

Moonshot Kimi 1 model

Kimi K2

kimi/kimi-k2-0711-preview

128K

$0.6

cached $0.15

$2.5

—

MiniMax 1 model

MiniMax M1

minimax/MiniMax-M1

$0.4

$2.1

Reasoning

Groq 2 models

Llama 3.3 70B

groq/llama-3.3-70b-versatile

131K

$0.59

$0.79

—

Llama 4 Maverick

groq/meta-llama/llama-4-maverick-17b-128e-instruct

131K

$0.2

$0.6

Vision

Together AI 3 models

Llama 4 Scout

together/meta-llama/Llama-4-Scout-17B-16E-Instruct

1.05M

$0.18

$0.59

Vision

DeepSeek V3

together/deepseek-ai/DeepSeek-V3

131K

$1.25

—

Qwen3 235B

together/Qwen/Qwen3-235B-A22B-fp8

131K

$0.2

$0.6

Reasoning

Fireworks AI 2 models

Llama 4 Maverick

fireworks/accounts/fireworks/models/llama4-maverick-instruct-basic

1.05M

$0.22

$0.88

Vision

DeepSeek R1

fireworks/accounts/fireworks/models/deepseek-r1

164K

Reasoning

Prices are provider list rates per 1M tokens from the July 2026 catalog, billed to you directly by each provider on your own keys. TokenRouter adds no markup. New and unlisted models pass straight through the gateway; check each provider for current rates.

Every model above, one base URL

Add your provider keys once, and switch between any of these models by changing a string, or let the router choose.

Start for free Models & routing docs