Skip to main content

Models & Pricing

Grida uses AI models across the editor for text generation, image generation, and image processing. This page documents the available models, their capabilities, and pricing.

Credits

Grida uses a credit system for AI operations. 1 credit = $0.0016 USD.

Credits provide a unified billing unit across all AI features regardless of the underlying model or provider.

Text Models

Text models power chat, content generation, summarization, and agentic features in the editor.

Models are organized into tiers based on capability and cost:

TierAudienceTypical use
nanoFree users / miscTitle generation, summarization, lightweight extraction
miniFree usersMain agent, sub-agent, multimodal tasks
proPaid usersMain agent, multimodal tasks
maxPaid usersHeaviest tasks, complex reasoning

Current Models

TierModelContext WindowMax OutputInput (per 1M tokens)Output (per 1M tokens)
nanoGPT-5 Nano (openai/gpt-5-nano)400K128K$0.05$0.40
miniGPT-5 Mini (openai/gpt-5-mini)400K128K$0.25$2.00
proClaude Sonnet 4 (anthropic/claude-sonnet-4)200K64K$3.00$15.00
maxClaude Opus 4 (anthropic/claude-opus-4)200K32K$15.00$75.00

All text models support multimodal inputs (text + images).

Cache Pricing

The pro and max tiers support prompt caching, which reduces cost for repeated context:

TierCache Read (per 1M tokens)Cache Write (per 1M tokens)
pro$0.30$3.75
max$1.50$18.75

Image Generation Models

Image models power the image generation features in the editor. Pricing is per image.

ModelProviderSpeedAvg. Cost/ImageCreditsStatus
GPT Image 1.5 (gpt-image-1.5)OpenAIMedium$0.02650Active
Flux Kontext Max (black-forest-labs/flux-kontext-max)ReplicateFastest$0.08050Active
Flux Pro 1.1 (black-forest-labs/flux-1.1-pro)ReplicateSlow$0.04025Active
GPT Image 1 (gpt-image-1)OpenAISlowest$0.09862Deprecated
Recraft V3 (recraft-ai/recraft-v3)ReplicateSlow$0.04025Deprecated
Flux Schnell (black-forest-labs/flux-schnell)ReplicateFastest$0.0032Deprecated

Image Sizes

Each model supports different output sizes. Default output is 1024x1024 (1:1).

ModelMin SizeMax SizeSupported Aspect Ratios
GPT Image 1.51024x10241536x15361:1, 2:3, 3:2
Flux Kontext Max1820x1820Any (freeform)
Flux Pro 1.1256x2561440x1440Any (freeform)

Image Tools

Image tools provide processing capabilities like upscaling and background removal.

Background Removal

ModelProviderAvg. Cost/ImageCredits
851 Labs Background Remover (851-labs/background-remover)Replicate$0.000481
Recraft Remove Background (recraft-ai/recraft-remove-background)Replicate$0.0107
Bria Remove Background (bria/remove-background)Replicate$0.01812

Upscale

ModelProviderAvg. Cost/ImageCredits
Real-ESRGAN (nightmareai/real-esrgan)Replicate$0.0022

Tier Selection Guide

Choose the right tier for your use case:

  • nano — Best for high-volume, low-complexity tasks where cost efficiency matters. Titles, summaries, simple extraction.
  • mini — Good balance of capability and cost. Suitable for most interactive agent tasks.
  • pro — Higher quality reasoning and generation. Use when output quality matters more than cost.
  • max — Maximum capability for the most demanding tasks. Complex multi-step reasoning, large codebases, nuanced analysis.

Pricing data sourced from models.dev. Prices reflect direct provider pricing and may change when models are updated.