Documentation Index
Fetch the complete documentation index at: /llms.txt
Use this file to discover all available pages before exploring further.

Skip to main content

Synth AI home page

Run 10-minute Quickstart
GitHub
Get Started
Get Started

Getting Started

Overview
Install and Authenticate
SynthClient
Get started
Quickstart
MCP Quickstart
Upgrading (0.11.2)

Managed Research

Introduction
Concepts
Synth Tag
Projects and Context
Runs and Evidence
GameBench
Usage and Plans
Scenarios
Beta Access
Launch Fields
Models and Harnesses
Runbooks and Work Modes
Providers and Credentials

Stack

Overview
Quickstart
Local-only
Connect Synth
Hosted ops
Cockpit
Goal & monitor
Stackd
Stack MCP
Configuration
Inference
Optimizers
Changelog

Research Factory

Research Factory
Factory Quickstart
Banking77 (Results)
Factories
Data Factory Flow

Prompt Optimization

Overview
Prompt Optimization
Hosted Optimizers

Infrastructure

In-Process Task App
Container Pools
Tunnels

SDK Reference

Sdk
SynthClient
ContainersClient
TunnelsClient
ContainerPoolsClient
Horizons Private compatibility

Research API

ResearchClient
account
Factories
factory_handles
factory_usage
Limits
Secrets
Projects
Project namespaces
swarms
ALPHA
swarm_readouts
experiments
knowledge
wiki
Models
Enums
Errors
AsyncResearchClient

Reference

Python SDK Reference
Python SDK Quickstart
MCP Tool Reference
Preflight and Errors
Outcome Rewards

Cookbooks

Eval Harness Improvement
Benchmark Improvement
Repo Review and PR
Checkpoint Branching
OpenCode Harness

synth_ai.research.enums

Research API

Enums

`synth_ai.research.enums`

Research launch enums (public names).

Enum	Values	Use
`ResearchWorkMode`	standard, …	`runs.create(work_mode=...)`
`ResearchHostKind`	container, …	Host selection on launch payloads

Was this page helpful?

Assistant

Responses are generated using AI and may contain mistakes.

Suggestions