Home › Free LLM API Models

Free LLM API Models

20 models in the catalog are free at list price — $0 per 1M input tokens and $0 per 1M output tokens. The free model with the largest context window is Thinking Machines: Inkling Small (free) at 1.04858M tokens. Free tiers can carry rate limits; check the model page for specs before building on one.

All free models, by context window

#ModelProviderInput /1MOutput /1MContext
1Thinking Machines: Inkling Small (free)thinkingmachinesFreeFree1,048,576
2Thinking Machines: Inkling (free)thinkingmachinesFreeFree1,048,576
3Google: Lyria 3 Pro PreviewgoogleFreeFree1,048,576
4Google: Lyria 3 Clip PreviewgoogleFreeFree1,048,576
5MiniMax: MiniMax M3.1 Flash (free)minimaxFreeFree1,000,000
6NVIDIA: Nemotron 3.5 Super VL (free)nvidiaFreeFree1,000,000
7NVIDIA: Nemotron 3.5 Lightning (free)nvidiaFreeFree1,000,000
8NVIDIA: Nemotron 3 Ultra (free)nvidiaFreeFree1,000,000
9Dots Studio: Dots3-Note Preview (free)dots-studioFreeFree512,000
10XingChen: Xing4.0 29B A4B (free)xingchen-agiFreeFree262,144
11Poolside: Laguna S 2.1 (free)poolsideFreeFree262,144
12Poolside: Laguna XS 2.1 (free)poolsideFreeFree262,144
13Google: Gemma 4 26B A4B (free)googleFreeFree262,144
14Google: Gemma 4 31B (free)googleFreeFree262,144
15NVIDIA: Nemotron 3 Super (free)nvidiaFreeFree262,144
16Cohere: North Mini Code (free)cohereFreeFree256,000
17NVIDIA: Nemotron 3 Nano Omni (free)nvidiaFreeFree256,000
18Free Models RouteropenrouterFreeFree200,000
19NVIDIA: Nemotron 3.5 Content Safety (free)nvidiaFreeFree128,000
20LiquidAI: LFM2.5-2.6B (free)liquidFreeFree65,536

Ranking from the October 21, 2026 snapshot, refreshed daily. How we collect prices