Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.factory.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Anthropic

ModelMultiplierBest for
Claude Opus 4.71x (2x after 4/30)Newest flagship, discounted during launch window
Claude Opus 4.62×Previous flagship, Max reasoning
Claude Opus 4.6 Fast12×Opus 4.6 tuned for faster responses
Claude Sonnet 4.61.2×Max reasoning at the Sonnet price point
Claude Opus 4.52×Complex reasoning, architecture
Claude Sonnet 4.51.2×Balanced quality/cost
Claude Haiku 4.50.4×Quick edits, routine work

​
OpenAI

ModelMultiplierBest for
GPT-5.52×Latest GPT flagship with Extra High reasoning
GPT-5.5 Fast5×GPT-5.5 on a priority service tier
GPT-5.5 Pro12×Higher-capability GPT-5.5 variant for research-heavy tasks
GPT-5.41×Large-context GPT model with Extra High reasoning
GPT-5.4 Fast2×Faster GPT-5.4 responses
GPT-5.4 Mini0.3×Cost-sensitive GPT work
GPT-5.3-Codex0.7×Advanced coding with Extra High reasoning
GPT-5.3-Codex Fast1.4×Faster GPT-5.3-Codex responses
GPT-5.20.7×General GPT work with Extra High reasoning
GPT-5.2-Codex0.7×Advanced coding with Extra High reasoning

​
Google

ModelMultiplierBest for
Gemini 3.1 Pro0.8×Research, analysis with newer Gemini generation
Gemini 3 Pro0.8×Research, analysis
Gemini 3 Flash0.2×Fast, cheap for high-volume tasks

​
Droid Core (Open Models)

ModelMultiplierBest for
GLM-5.10.55×Newer open-source GLM model when you want stronger quality in Droid Core
Kimi K2.60.4×Cost-sensitive work, supports images, optional High reasoning
Kimi K2.50.25×Cost-sensitive work, supports images
MiniMax M2.70.12×Cheapest option with reasoning support

​
Custom models

Configure custom models through Bring Your Own Key (BYOK). Custom models use the custom:<alias> model ID format. Context windows, reasoning support, and multipliers depend on the configured provider and model.