Skip to main content

Your Privacy Choices

Copyright © 2026 NVIDIA Corporation

Models

Deploy and scale models on your GPU infrastructure of choice with NVIDIA NIM inference microservices

Optimized by NVIDIA Launch from Hugging FaceBeta

Filters (2)

Free Endpoint2

Partner Endpoint2

Download Available2

Use Case

Image-to-Text1Drug Discovery0Retrieval Augmented Generation0Speech-to-Text0Image Generation0

Inference Providers

OpenRouter2Together AI2GMI Cloud2Deepinfra1Bitdeer1

Publisher

Moonshotai1Thinkingmachines1NVIDIA0Meta0Google0

NIM Container GPUs

H100 80GB HBM30A100 SXM4 80GB0L40S0A10G0B2000

Labels (2)

MultimodalReasoning

2 models

Sort By

Thinkingmachines

DownloadableFree Endpoint

inkling

Inkling is a multimodal (text + image) reasoning model from Thinking Machines — a Mamba-hybrid, 256-expert Mixture-of-Experts architecture with tool use and switchable reasoning.

reasoning image-to-text multimodal

Last updated on July 16, 2026

Items per page

of 1 pages

DownloadableFree Endpoint

kimi-k2.6

1T multimodal MoE for long-horizon coding, agentic tool use, and image/video understanding.

Mixture-of-Experts Reasoning Image-to-Text

16M API calls in the last 30 days

Last updated on May 1, 2026