MiroMind · emploidai Marketplace
emploidai Marketplace
Add-onsAppletsPlugins
Search tools, teams, and capabilitiesPublish
MarketplacePluginsMiroMind
Plugin
Limited listing

MiroMind

by miromind · v0.0.1

MiroMind deep-research models. Long-running streaming research agents with reasoning, web search and inline citations.

65 installsUpdated May 19, 2026
Publisher information is incomplete

This community listing does not yet include every recommended support, privacy, pricing, and permission disclosure. Review the available package permissions before installing.

Capabilities

Models

Available inside your emploidai workspace after installation.

Data sources

Available inside your emploidai workspace after installation.

Category

model

Version

0.0.1miromind

Requirements

Maximum memory 256MB

Pricing

Not disclosed by publisher

Security & access

Review before installing

CompatibleRequires emploidai 1.0.0+

Permissions

No permissions were declared by the publisher.

Dependencies

No additional dependencies

Resources

Privacy policy
emploidai Marketplace

Discover capabilities. Review access. Install inside your workspace.

DocumentationSecuritySupportPrivacyTerms

MiroMind for Dify

Type: Model plugin · Author: miromind · Status: MVP, private preview

Brings MiroMind deep-research models to Dify. Two models are shipped:

ModelContextMax outputPricing (USD per 1M tokens)
mirothinker-1-7-deepresearch256k16k$4.00 in / $25.00 out
mirothinker-1-7-deepresearch-mini256k16k$1.25 in / $10.00 out

What's different from a normal chat model

These are deep-research agents, not chatbots:

  • Long-running. A single query routinely takes minutes to tens of minutes. The plugin opens a streaming connection with no read timeout; MiroMind's API sends regular SSE heartbeats so intermediaries don't drop the connection.
  • Single-turn. Each request is an independent research task, not a chat with memory. The plugin collapses Dify's conversation history into the latest user message; prior turns are not sent to the model.
  • Reasoning visible. Research progress (reasoning_text and agent_summary events) is streamed into Dify's reasoning channel — users see the model's thinking in the collapsible reasoning UI before the final answer arrives.
  • Inline citations. Citations are emitted as markdown links inside the answer text and render natively in the Dify chat UI.

Setup

  1. In Dify → Settings → Model Providers, install this plugin.
  2. Provide an API key from platform.miromind.ai.
  3. Optionally override the base URL (defaults to https://api.miromind.ai/v1).
  4. Pick a model in any chatbot / workflow node.

API endpoint

The plugin calls POST /v1/responses (OpenAI Responses API shape, with MiroMind extensions) and streams SSE events back. Resume / replay support (?after=<sequence_number>) is on the platform side; transparent client-side resume is planned for a follow-up release.

License

Apache-2.0 (TBD).