Question 1

Does Wysor train AI models on my data?

Accepted Answer

No. No provider on Wysor uses your conversations for training. This applies to all 15+ models including OpenAI, Anthropic, Google, and xAI.

Question 2

What is zero data retention and why does it matter?

Accepted Answer

Zero data retention means the AI provider processes your message and immediately forgets it. No copy is saved on their servers. Most models on Wysor have this by default. For OpenAI models, Wysor explicitly requests that nothing is stored on every single message.

Question 3

Is zero data retention more important than EU hosting?

Accepted Answer

Yes. An AI provider hosted in the EU that stores your data for 30 days is less private than a provider hosted elsewhere that deletes your data immediately. Server location matters, but what happens to your data matters more.

Question 4

Which AI models on Wysor have zero data retention?

Accepted Answer

Every model Wysor offers runs with zero data retention. Anthropic (Claude), Google (Gemini) and the full OpenAI GPT-5.6 range (via AWS Bedrock) process your prompts and immediately discard them, as do the open-weight models (DeepSeek, GLM, Kimi, Qwen, MiniMax) on Fireworks AI. Wysor no longer sends any prompt to OpenAI’s direct API, so the 30-day safety-retention window that applied to the smaller OpenAI models no longer applies to anything in the catalogue.

Question 5

How many AI models does Wysor offer?

Accepted Answer

Wysor offers 15+ AI models from OpenAI, Anthropic, Google, xAI, and open-source providers. All models include web search and none train on your data.

modelsPage.tableModel	modelsPage.tableIntelligence	modelsPage.tableCost	modelsPage.tableYourData	modelsPage.tableBestFor
Claude Opus 5modelsPage.tablePopular Anthropic	61	2x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.claude-opus-5
Grok 4.6 xAI	61	1x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.grok-4-6
GPT-5.6 SolmodelsPage.tablePopular OpenAI	59	3x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.gpt-5-6-sol
Gemini 3.8 Flash Google	59	1x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.gemini-3-8-flash
Gemini 3.1 PromodelsPage.tablePopular Google	57	1.5x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.gemini-3-1-pro
Kimi K3 Open Source	57	1.5x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.kimi-k3
Gemini 3.5 Flash Google	55.3	1x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.gemini-3-5-flash
GPT-5.6 Terra OpenAI	55	1.5x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.gpt-5-6-terra
Claude Sonnet 5modelsPage.tablePopular Anthropic	53	1.5x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.claude-sonnet-5
DeepSeek V4 Pro Open Source	52	1x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.deepseek-v4-pro
GPT-5.6 Luna OpenAI	51	1x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.gpt-5-6-luna
GLM-5.3 Open Source	51	1x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.glm-5-3
Qwen3.8 Max Open Source	46	1x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.qwen3-8-max
MiniMax M3 Open Source	44	0.5x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.minimax-m3
Claude 4.5 Haiku Anthropic	37	1x	modelsPage.tableNotStoredNotTrained	modelsPage.modelDesc.claude-4-5-haiku

modelsPage.title

modelsPage.barChartTitle

modelsPage.zdrTitle

modelsPage.tipsTitle

modelsPage.tip1Title

modelsPage.tip2Title

modelsPage.tip3Title

modelsPage.tip4Title

modelsPage.ctaTitle