Cette page est affichée en anglais. Une traduction française est en cours.
ConfigsMay 8, 2026·5 min de lecture

Cherry Studio Custom Models — BYOK Any LLM Provider

Cherry Studio Custom Models adds any OpenAI-compatible endpoint — proxy, local, or third-party. Mix Claude, GPT, Gemini, DeepSeek, Ollama side-by-side.

Cherry Studio
Cherry Studio · Community
Prêt pour agents

Cet actif peut être lu et installé directement par les agents

TokRepo expose une commande CLI universelle, un contrat d'installation, le metadata JSON, un plan selon l'adaptateur et le contenu raw pour aider les agents à juger l'adaptation, le risque et les prochaines actions.

Needs Confirmation · 66/100Policy : confirmer
Surface agent
Tout agent MCP/CLI
Type
Skill
Installation
Single
Confiance
Confiance : New
Point d'entrée
Asset
Commande CLI universelle
npx tokrepo install 121c2d1e-678a-4ea4-8799-70391f0a11de
contrat d'installationJSON metadataplan adaptateurcontenu raw
Introduction

Cherry Studio Custom Models support lets you add any OpenAI-compatible endpoint as a model — your own LiteLLM proxy, an Ollama instance, OpenRouter, DeepSeek, anything. Multiple models at once, swap mid-conversation, or run side-by-side comparisons. Best for: power users who want one interface across all their LLM providers, including local models. Works with: Cherry Studio 1.x, any OpenAI-compatible endpoint. Setup time: 2 minutes per model.

Add a custom provider

Settings → Models → ➕ Add Provider:

Provider Type: OpenAI Compatible
Name: My LiteLLM Proxy
Base URL: https://litellm.acme.internal/v1
API Key: sk-team-acme-xyz

Cherry Studio fetches the model list via /v1/models. Select the ones you want exposed.

Common providers

Provider Base URL Models
OpenRouter https://openrouter.ai/api/v1 300+
Together AI https://api.together.xyz/v1 200+
Groq https://api.groq.com/openai/v1 Fast Llama, Mixtral
DeepSeek https://api.deepseek.com/v1 DeepSeek-V3, R1, Coder
Ollama (local) http://localhost:11434/v1 Anything you've pulled
LM Studio (local) http://localhost:1234/v1 GGUF models
Your LiteLLM Proxy https://your-proxy.com/v1 Whatever it routes

Side-by-side comparison

In a chat, click the model selector → ⊕ Add Model. Two (or more) models respond to the same prompt simultaneously. Useful for picking the right model for a workload.

Switch mid-conversation

Click the model name on any message → switch the model used for the next message. Great for "use Sonnet for the hard reasoning, switch to Haiku for the boilerplate replies."

Config persistence

Provider configs are stored in ~/Library/Application Support/CherryStudio/config.json (encrypted with the OS keychain). Sync via the optional Cherry Studio Sync (encrypted to your passphrase) if you want them across devices.

FAQ

Q: Does it work with Anthropic Claude directly? A: Cherry Studio supports Anthropic's native API as a built-in provider type (not OpenAI-compatible). Add via Settings → Models → ➕ Add Provider → Anthropic. Same flow as the others.

Q: Can I use a self-hosted LLM? A: Yes — point to your local Ollama (port 11434) or LM Studio (port 1234) for fully local inference. The OpenAI-compatible adapter handles tool use and streaming for both.

Q: How are API keys stored? A: Locally only. Cherry Studio uses the OS keychain (macOS Keychain / Windows Credential Manager / Linux Secret Service). Keys never leave your device unless you opt into Cherry Studio Sync (encrypted).

Quick Use

  1. Open Cherry Studio → Settings → Models → ➕ Add Provider
  2. Pick OpenAI Compatible, paste base URL and API key
  3. Cherry Studio auto-fetches models via /v1/models — select the ones you want

Intro

Cherry Studio Custom Models support lets you add any OpenAI-compatible endpoint as a model — your own LiteLLM proxy, an Ollama instance, OpenRouter, DeepSeek, anything. Multiple models at once, swap mid-conversation, or run side-by-side comparisons. Best for: power users who want one interface across all their LLM providers, including local models. Works with: Cherry Studio 1.x, any OpenAI-compatible endpoint. Setup time: 2 minutes per model.

Add a custom provider

Settings → Models → ➕ Add Provider:

Provider Type: OpenAI Compatible
Name: My LiteLLM Proxy
Base URL: https://litellm.acme.internal/v1
API Key: sk-team-acme-xyz

Cherry Studio fetches the model list via /v1/models. Select the ones you want exposed.

Common providers

Provider Base URL Models
OpenRouter https://openrouter.ai/api/v1 300+
Together AI https://api.together.xyz/v1 200+
Groq https://api.groq.com/openai/v1 Fast Llama, Mixtral
DeepSeek https://api.deepseek.com/v1 DeepSeek-V3, R1, Coder
Ollama (local) http://localhost:11434/v1 Anything you've pulled
LM Studio (local) http://localhost:1234/v1 GGUF models
Your LiteLLM Proxy https://your-proxy.com/v1 Whatever it routes

Side-by-side comparison

In a chat, click the model selector → ⊕ Add Model. Two (or more) models respond to the same prompt simultaneously. Useful for picking the right model for a workload.

Switch mid-conversation

Click the model name on any message → switch the model used for the next message. Great for "use Sonnet for the hard reasoning, switch to Haiku for the boilerplate replies."

Config persistence

Provider configs are stored in ~/Library/Application Support/CherryStudio/config.json (encrypted with the OS keychain). Sync via the optional Cherry Studio Sync (encrypted to your passphrase) if you want them across devices.

FAQ

Q: Does it work with Anthropic Claude directly? A: Cherry Studio supports Anthropic's native API as a built-in provider type (not OpenAI-compatible). Add via Settings → Models → ➕ Add Provider → Anthropic. Same flow as the others.

Q: Can I use a self-hosted LLM? A: Yes — point to your local Ollama (port 11434) or LM Studio (port 1234) for fully local inference. The OpenAI-compatible adapter handles tool use and streaming for both.

Q: How are API keys stored? A: Locally only. Cherry Studio uses the OS keychain (macOS Keychain / Windows Credential Manager / Linux Secret Service). Keys never leave your device unless you opt into Cherry Studio Sync (encrypted).

Source & Thanks

Built by kangfenmao. Licensed under Apache-2.0.

CherryHQ/cherry-studio — ⭐ 18,000+

🙏

Source et remerciements

Built by kangfenmao. Licensed under Apache-2.0.

CherryHQ/cherry-studio — ⭐ 18,000+

Fil de discussion

Connectez-vous pour rejoindre la discussion.
Aucun commentaire pour l'instant. Soyez le premier à partager votre avis.

Actifs similaires

Cherry Studio Knowledge Base — Local RAG with 50+ Formats

Cherry Studio Knowledge Base ingests PDFs, Office docs, Markdown into a local vector index. Query offline, BYOK any LLM. Data stays on your machine.

Knowledge
Cherry Studio

Unsloth — 2x Faster Local LLM Training & Inference

Unsloth is a unified local interface for running and training AI models. 58.7K+ GitHub stars. 2x faster training with 70% less VRAM across 500+ models including Qwen, DeepSeek, Llama, Gemma. Web UI wi

Configs
AI Open Source

Cherry Studio — AI Productivity Desktop with 300+ Assistants

Desktop AI client with smart chat, autonomous agents, and unified access to frontier LLMs across providers.

Scripts
Cherry Studio

Guardrails AI — Validate LLM Outputs in Production

Add validation and guardrails to any LLM output. Guardrails AI checks for hallucination, toxicity, PII leakage, and format compliance with 50+ built-in validators.

Workflows
Agent Toolkit
◈Accueil🔍Rechercher👤Moi