All AI models · Perplexity
Perplexity model list
All 3 notable Perplexity models tracked in 2026, out of 279 across every lab. Sorted newest first, with context length and capabilities. Data from the open models.dev database.
3
Models
1
Reasoning
2
Vision
0
Open weights
| Model | Context | Capabilities |
|---|---|---|
|
Sonar
perplexity/sonar
Fast web-grounded Sonar for current answers, citations, and lightweight retrieval
|
128K | |
|
Sonar Pro
perplexity/sonar-pro
Deeper Sonar search model with broader retrieval and stronger synthesis
|
200K | Vision |
|
Sonar Reasoning Pro
perplexity/sonar-reasoning-pro
Web-grounded Sonar for multi-step research questions that need cited reasoning
|
128K | Reasoning Vision |
Which of these can I call for free?
This page lists what exists, not what is free. For the models you can call right now with one free key and no credit card, see the live API catalog or read the free LLM API guide. Open-weight models can also be run on your own GPU, and the VRAM calculator tells you whether yours is big enough.
Model lists by lab
OpenAI (56)
Google (43)
Alibaba (Qwen) (33)
Anthropic (23)
NVIDIA (20)
Mistral (18)
Zhipu AI (14)
Cohere (14)
MiniMax (7)
Moonshot AI (7)
xAI (6)
Xiaomi (6)
DeepSeek (5)
Meta (4)
Poolside (4)
DeepReinforce (4)
Perplexity (3)
StepFun (3)
Sakana AI (2)
Tencent (2)
Sarvam AI (2)
Microsoft (1)
Thinking Machines (1)
Meituan (1)
Model metadata sourced from the open models.dev database. See the full 279-model index for a filterable view across every lab.