Routed Models

Frontier models the gateway routes to an upstream provider, so one key and one OpenAI-compatible endpoint cover them as well. Prices are the upstream list prices per million tokens.

Kimi K3

moonshotai/kimi-k3

Context
262K
Input / 1M
$3
Cached / 1M
$0.30
Output / 1M
$15

Moonshot open trillion-scale reasoning model for coding, research, and long agent runs.

Muse Spark 1.2

meta/muse-spark-1.2

Context
1M
Input / 1M
$1.25
Cached / 1M
$0.15
Output / 1M
$4.25

Meta reasoning model for complex agent tasks, with a one million token context.

Muse Glimmer 30B

meta/muse-glimmer-30b

Context
131K
Input / 1M
$0.35
Cached / 1M
$0.04
Output / 1M
$1.50

Compact open Meta model distilled from Muse Spark for agents on local hardware.

Gemini 3.7 Flash

google/gemini-3.7-flash

Context
1M
Input / 1M
$0.75
Cached / 1M
$0.075
Output / 1M
$3.75

Multimodal Google model for quick agent loops, coding, and multi-step reasoning.

Gemini 3.5 Flash Lite

google/gemini-3.5-flash-lite

Context
1M
Input / 1M
$0.30
Cached / 1M
$0.03
Output / 1M
$2.50

Low-cost Google model tuned for focused subagent steps inside larger workflows.

Gemini Robotics-ER 2 (preview)

google/gemini-robotics-er-2-preview

Context
1M
Input / 1M
$0.30
Cached / 1M
$0.03
Output / 1M
$2.50

Google embodied reasoning preview for spatial understanding, planning, and robot control.

Gemma 4 31B Instruct

google/gemma-4-31b-it

Context
131K
Input / 1M
$0.15
Cached / 1M
$0.05
Output / 1M
$0.40

Dense open Google model for text and images, with an optional thinking mode.

Gemma 4 26B-A4B Instruct

google/gemma-4-26b-a4b-it

Context
131K
Input / 1M
$0.10
Cached / 1M
$0.05
Output / 1M
$0.30

Open Google mixture-of-experts model: 26B total weights, about 4B active per token.

MiniMax M3

minimax/minimax-m3

Context
1M
Input / 1M
$0.30
Cached / 1M
$0.06
Output / 1M
$1.20

MiniMax multimodal model for long-horizon agent work, coding, and very long contexts.

Showing 9 of 9 models

Get an API key. Ship vision today.