AI Models

Compare and switch between different AI models. Select one to use in the playground.

Nemotron 3 Ultra

NVIDIA

NVIDIA's most powerful Nemotron, 550B MoE model — free tier on OpenRouter.

Free550B MoEFastReasoning

Latency

0.6s

Context

128K

Pricing

Free

Gemma 4 31B

Google

Google's latest Gemma 4 flagship with 1M context window — free tier on OpenRouter.

Free1M ContextLatestReasoning

Latency

0.8s

Context

1M

Pricing

Free

Gemma 4 26B

Google

Google's MoE Gemma 4 model — fast, efficient, free on OpenRouter.

FreeMoE1M ContextEfficient

Latency

0.6s

Context

1M

Pricing

Free

Gemma 3 12B

Google

Google's well-rounded open model with strong quality.

FreeFastOpen SourceBalanced

Latency

0.5s

Context

128K

Pricing

Free

Llama 3.1 70B

Meta

Meta's open-source large language model, deployable on-premise.

Open SourceCodeReasoningOn-premise

Latency

1.5s

Context

128K

Pricing

Free (Open Source)

Mistral Large

Mistral

Mistral's flagship model with strong multilingual and code capabilities.

MultilingualCodeReasoningOpen Source

Latency

1.1s

Context

128K

Pricing

$8 / 1M tokens

Active: Nemotron 3 Ultra

This model will be used in the playground and API requests