← Back to Home

Available Models

All AI models available on the HeySalad platform

Gemini 3 Series (Latest)

Google's latest models with extended thinking capabilities

gemini-3.1-pro-preview

Google

Available
  • Extended Thinking (95 tokens)
  • Balanced performance
  • Latest features
Speed
Medium (~5s)
Quality
Excellent

gemini-3-flash-preview

Google

Available
  • Extended Thinking (65 tokens)
  • Fastest response
  • High efficiency
Speed
Fast (~3s)
Quality
Very Good

gemini-3-pro-preview

Google

Available
  • Extended Thinking (263 tokens)
  • Deep reasoning
  • Best quality
Speed
Slow (~6s)
Quality
Best

Gemini 2.5 Series

Stable, production-ready Gemini models

gemini-2.5-flash

Google

Available
  • Fast responses
  • Production ready
  • Balanced cost
Speed
Fast
Quality
Very Good

gemini-2.5-flash-lite

Google

Available
  • Lightest model
  • Cheapest option
  • High throughput
Speed
Fastest
Quality
Good

gemini-2.5-pro

Google

Available
  • Extended thinking
  • High quality
  • Production ready
Speed
Medium
Quality
Excellent

gemini-2.5-flash-image

Google

Available
  • Vision support
  • Image analysis
  • Text generation
Speed
Fast
Quality
Very Good

OpenAI Models

GPT models from OpenAI

gpt-4-turbo

OpenAI

Available
  • 128K context
  • Vision support
  • Latest GPT-4
Speed
Medium
Quality
Excellent

gpt-4

OpenAI

Available
  • Advanced reasoning
  • 8K context
  • Reliable
Speed
Slow
Quality
Excellent

gpt-3.5-turbo

OpenAI

Available
  • Fast responses
  • 16K context
  • Cost effective
Speed
Fast
Quality
Good

Anthropic Models

Claude models from Anthropic

claude-opus-4

Anthropic

Available
  • 200K context
  • Best reasoning
  • Extended thinking
Speed
Slow
Quality
Best

claude-sonnet-4

Anthropic

Available
  • 200K context
  • Balanced
  • Fast and capable
Speed
Fast
Quality
Excellent

claude-haiku-4

Anthropic

Available
  • 200K context
  • Fastest
  • Cost optimized
Speed
Fastest
Quality
Good

Hugging Face Models

Open source models via Hugging Face API

meta-llama/Llama-3.2-3B-Instruct

Hugging Face

Available
  • Open source
  • Cost effective
  • Self-hostable
Speed
Fast
Quality
Good

mistralai/Mistral-7B-Instruct

Hugging Face

Available
  • Open source
  • 32K context
  • Efficient
Speed
Fast
Quality
Very Good