Skip to main content

Deep Infra

Inference platform

Deep Infra is a Inference platform provider with 11 models across 11 catalog listings.

Models
11
Model families
4
Output coverage
Text
Latest release
2026-07-16
SDK
@ai-sdk/deepinfra
Environment variables
DEEPINFRA_API_KEY

Models

11 of 11 model offerings

Cache write
Kimi K3Moonshot AImoonshotai/Kimi-K3Openai chat1M131KText, ImageText-
Kimi K2.7 CodeMoonshot AImoonshotai/Kimi-K2.7-CodeOpenai chat262K262KText, Image, VideoText-
Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIAnvidia/Nemotron-3-Nano-Omni-30B-A3B-ReasoningOpenai chat262K66KText, Image, Video, AudioText--
Kimi K2.6Moonshot AImoonshotai/Kimi-K2.6Openai chat262K16KText, Image, VideoText-
Gemma 4 26B A4B ITGooglegoogle/gemma-4-26B-A4B-itOpenai chat262K33KText, ImageText--
Gemma 4 31B ITGooglegoogle/gemma-4-31B-itOpenai chat262K33KText, Image, VideoText--
Kimi K2.5Moonshot AImoonshotai/Kimi-K2.5Openai chat262K33KText, Image, VideoText-
Nemotron 3 Nano 30B A3BNVIDIAnvidia/Nemotron-3-Nano-30B-A3BOpenai chat262K262KTextText-
gpt-oss-120bOpenAIopenai/gpt-oss-120bOpenai chat131K16KTextText--
gpt-oss-20bOpenAIopenai/gpt-oss-20bOpenai chat131K16KTextText--
Llama 3.3 Nemotron Super 49B v1.5NVIDIAnvidia/Llama-3.3-Nemotron-Super-49B-v1.5Openai chat131K131KTextText--

Page 1 of 1

Other providers