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

# Agent compatibility

> Which AI agents, editors and chat apps work with the Fortytwo API, what each one needs, and where to find its setup guide.

Fortytwo tests these agents and apps with the Fortytwo API: chat, streaming and tool calls. Any other tool that lets you add an OpenAI-compatible Chat Completions provider can work too: use its own guide and the settings below. Choose Chat Completions, not the Responses API.

> [!NOTE]
>
> **Before you start:** create a Fortytwo API key and buy credits — see [Quickstart, step 1](https://docs.fortytwo.network/docs/quickstart.md#1-create-an-api-key). Give each agent its own key with a **Spend limit**.
>
> | Setting | Value |
> | - | - |
> | Base URL | `https://api.fortytwo.network/v1` |
> | Model ID | `qwen/qwen3.8-27b` |

| Status | Meaning |
| - | - |
| Works | Works with the settings above |
| Works with a setting | Works after one extra setting, named in the notes |
| Needs a bridge | Uses another API; works through LiteLLM |
| Not compatible | Cannot connect to an OpenAI-compatible Chat Completions provider |
| Not tested | Has a custom provider setting; not tested yet |

The model returns its reasoning in the `reasoning` field. Agents that read only `reasoning_content` work, but do not show the reasoning.

**Terminal agents**

| Agent | Status | Notes | Setup guide |
| - | - | - | - |
| Hermes Agent | Works | Run `hermes model`, choose **Custom endpoint**, then **Chat Completions** | [Fortytwo guide](https://docs.fortytwo.network/agents/hermes-agent.md#terminal) |
| pi | Works | Provider in `~/.pi/agent/models.json` with `"api": "openai-completions"`. Set `/thinking` to **medium**; pi starts with thinking off | [Fortytwo guide](https://docs.fortytwo.network/agents/pi.md) |
| Kilo CLI | Works with a setting | Set `limit.context` and `limit.output` for the model; with a small context, compaction fails | [Fortytwo guide](https://docs.fortytwo.network/agents/kilo-cli.md) |
| OpenCode | Works | Same steps as Kilo CLI | [Fortytwo guide](https://docs.fortytwo.network/agents/kilo-cli.md#opencode) |
| OpenClaw | Works with a setting | `openclaw onboard` → **Custom Provider**. Then raise `maxTokens` to 16384 and `contextWindow` to 262144 in `~/.openclaw/openclaw.json`: the wizard sets 4096 and 128000 | [Official guide ](https://docs.openclaw.ai/concepts/model-providers/custom-providers) |
| Crush | Works | Provider with `type: openai-compat` | [Official guide ](https://github.com/charmbracelet/crush#openai-compatible-apis) |
| Qwen Code | Works | OpenAI-compatible provider | [Official guide ](https://qwenlm.github.io/qwen-code-docs/en/users/configuration/model-providers/#openai-compatible-providers-openai) |
| Goose | Works | Custom provider. The built-in OpenAI provider takes the host `https://api.fortytwo.network` and the path `v1/chat/completions` separately | [Official guide ](https://goose-docs.ai/docs/getting-started/providers/#using-custom-openai-endpoints) |
| Cline CLI | Works | `cline auth -p openai -b https://api.fortytwo.network/v1 -k <key> -m qwen/qwen3.8-27b` | [Official guide ](https://docs.cline.bot/cli/cli-reference#auth-options-provider) |
| Continue CLI | Works | Uses the same `~/.continue/config.yaml` as the extension | [Official guide ](https://docs.continue.dev/cli/configuration) |
| Aider | Works with a setting | Use `--edit-format diff` and the model `openai/qwen/qwen3.8-27b`. Does not show reasoning | [Official guide ](https://aider.chat/docs/llms/openai-compat.html) |
| Codex | Needs a bridge | Uses only the Responses API | [Fortytwo guide](https://docs.fortytwo.network/agents/codex.md) |
| Claude Code | Not compatible | Uses only the Anthropic Messages API | — |
| Gemini CLI | Not compatible | Uses only the Gemini API | — |
| Factory Droid | Not tested | Custom model with `provider: "generic-chat-completion-api"` | [Official guide ](https://docs.factory.com/model-independence/byok) |
| Mistral Vibe | Not tested | Provider with `api_base` and `api_key_env_var` | [Official guide ](https://docs.mistral.ai/vibe/code/cli/configuration-reference) |
| Junie CLI | Not tested | `apiType: "OpenAICompletion"`; `baseUrl` takes the full `/v1/chat/completions` URL | [Official guide ](https://junie.jetbrains.com/docs/custom-llm-models.html) |
| Amp | Not tested | **Custom URL** connection with the `chat-completions` format | [Official guide ](https://ampcode.com/docs/customize/model-routing#custom-url-connections) |
| OpenHands | Not tested | Model `openai/qwen/qwen3.8-27b` | [Official guide ](https://docs.openhands.dev/openhands/usage/llms/openai-llms#using-openai-compatible-endpoints) |

**Editors**

| Agent | Status | Notes | Setup guide |
| - | - | - | - |
| Kilo Code | Works | **Settings** → **Providers** → **Custom provider** | [Fortytwo guide](https://docs.fortytwo.network/agents/kilo-code.md) |
| Cline | Works | **API Provider**: **OpenAI Compatible**. Set **Reasoning Effort** to **Medium**: the default **None** turns reasoning off | [Fortytwo guide](https://docs.fortytwo.network/agents/cline.md) |
| Continue | Works | `provider: openai` and `apiBase` in `config.yaml` | [Official guide ](https://docs.continue.dev/customize/model-providers/top-level/openai#openai-api-compatible-providers) |
| Zed | Works | OpenAI-compatible provider | [Official guide ](https://zed.dev/docs/ai/use-api-access#openai-compatible) |
| Roo Code | Works | Does not show reasoning. The project is archived | [Official guide ](https://roocodeinc.github.io/Roo-Code/providers/openai-compatible/) |
| VS Code (GitHub Copilot) | Not tested | **Custom Endpoint** model with **Chat Completions** | [Official guide ](https://code.visualstudio.com/docs/agent-customization/language-models#_add-a-custom-endpoint-model) |
| JetBrains AI Assistant | Not tested | OpenAI-compatible provider | [Official guide ](https://www.jetbrains.com/help/ai-assistant/use-custom-models.html) |
| Cursor | Not tested | No official guide for a custom base URL | — |
| Windsurf | Not compatible | No custom provider setting | — |

**Desktop apps**

| App | Status | Notes | Setup guide |
| - | - | - | - |
| Hermes Agent | Works | **Settings** → **Providers** → **Custom Endpoints** | [Fortytwo guide](https://docs.fortytwo.network/agents/hermes-agent.md) |
| Jan | Works | **Settings** → **Model Providers** → add a provider | [Official guide ](https://docs.jan.ai/desktop/remote-models/custom-endpoint) |
| Cherry Studio | Works | API host `https://api.fortytwo.network`, without `/v1` | [Official guide ](https://docs.cherryai.com.cn/docs/en-us/pre-basic/providers/zi-ding-yi-fu-wu-shang) |
| AnythingLLM | Works | LLM provider **Generic OpenAI** | [Official guide ](https://docs.anythingllm.com/setup/llm-configuration/cloud/openai-generic) |
| Goose | Works | Shares its providers with Goose in the terminal | [Official guide ](https://goose-docs.ai/docs/getting-started/providers/#using-custom-openai-endpoints) |

**Web chat apps**

| App | Status | Notes | Setup guide |
| - | - | - | - |
| Open WebUI | Works | OpenAI-compatible connection | [Official guide ](https://docs.openwebui.com/getting-started/quick-start/connect-a-provider/starting-with-openai-compatible/) |
| LibreChat | Works | Custom endpoint in `librechat.yaml` | [Official guide ](https://www.librechat.ai/docs/quick_start/custom_endpoints) |
| LobeHub | Works | Self-hosted: `OPENAI_PROXY_URL` and `OPENAI_MODEL_LIST` | [Official guide ](https://lobehub.com/docs/self-hosting/environment-variables/model-provider#openai-proxy-url) |
| SillyTavern | Works | Chat Completion source **Custom (OpenAI-compatible)** | [Official guide ](https://docs.sillytavern.app/usage/api-connections/openai/#custom-openai-compatible-endpoint) |
| NextChat | Works | `BASE_URL` and `CUSTOM_MODELS` | [Official guide ](https://github.com/ChatGPTNextWeb/NextChat#base_url-optional) |
| AnythingLLM (Docker) | Works | LLM provider **Generic OpenAI** | [Official guide ](https://docs.anythingllm.com/setup/llm-configuration/cloud/openai-generic) |

***

Something not working with your agent, or using one we have not listed? Tell us in the platform: [Feedback](https://docs.fortytwo.network/docs/support.md#ideas-feedback-and-small-bugs) for ideas and small bugs, [Contact support](https://docs.fortytwo.network/docs/support.md#support-account-api-billing-credits-errors) if it blocks your work, or [Report a security issue](https://docs.fortytwo.network/docs/support.md#security-issue).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.