> ## Documentation Index > Fetch the complete documentation index at: https://fireworks.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # Release Notes > What's new in each Nexus CLI release: new harnesses and commands, fixes, and what was deprecated or removed. Nexus CLI releases are listed newest first. To update, run: ```bash wrap theme={null} nexus update ``` Then restart your harnesses so they load the new settings. * Run `nexus --version` to see which version you have. * For every deprecated or removed command in one table, see [Deprecated and removed](https://fireworks.ai/docs/nexus/cli-reference.md#deprecated-and-removed). * Versions before 0.10.0 shipped as FireConnect, with the `fireconnect` command, and Nexus Auto was called FireRouter. Their notes use today's command names. See [Upgrading from FireConnect](https://fireworks.ai/docs/nexus/cli.md#upgrading-from-fireconnect). **0.10.1** October 22, 2026 **New** * **Bring your own Anthropic and OpenAI keys to Nexus Auto** with `nexus auto`. * **`nexus auto` shows every key**, both your saved keys and your account's Provider Keys. * **`nexus status` lists installed harnesses** as well as connected ones. * **`nexus update` updates Nexus CLI.** **Improved** * Connecting a harness that isn't installed stops and writes nothing. * Claude Code: the Artifact tool, `/model` picker, and status line work better. * Nexus CLI prompts for keychain access less often. * Sign-in no longer tells account members that they aren't members. * `nexus logout` also clears the Anthropic and OpenAI keys saved on this machine. * `nexus update` no longer modifies a `~/.nexus` directory that Nexus CLI didn't create. * The installer now requires Node.js 22 or later. **Deprecated and removed** * Removed `nexus claude live`. * Removed the Anthropic and OpenAI key flags from harness commands. * Deprecated `nexus harnesses` in favor of `nexus status`. * Deprecated the Anthropic and OpenAI key flags on `nexus configure`. - Bring your own keys: run `nexus auto --anthropic-api-key` or `nexus auto --openai-api-key`. Each prompts for the key, so it stays out of your shell history. Then run your harness command again. See [Nexus Auto](https://fireworks.ai/docs/nexus/auto.md#nexus-cli). - `nexus auto` shows the keys saved on this machine next to the [Provider Keys](https://fireworks.ai/docs/nexus/provider-keys.md) on your Fireworks account. Provider Keys work in every harness, and a key saved on your machine wins for its provider. - `nexus upgrade` still works. - Uninstalled harness: connect stops with an explanation. Pass `--skip-install-check` if the tool lives somewhere Nexus CLI doesn't look. - Claude Code: the Artifact tool shows up, a Claude Code older than 2.1.242 gets a warning instead of an empty `/model` picker, and the status line names the model that served the latest turn. Run `nexus claude` again to refresh the picker. - Keychain: Nexus CLI reads your keychain only when a command needs your key. - Safer updates: if `~/.nexus` wasn't created by Nexus CLI, `nexus update` stops before changing anything. - `nexus claude live` replacement: the Claude Code status line shows the model and spend on every turn, and `nexus claude usage` gives the full breakdown. - Removed `--anthropic-api-key` and `--openai-api-key` from harness commands such as `nexus opencode`. Save keys with `nexus auto` instead. - Deprecated `--anthropic-api-key`, `--openai-api-key`, and `--unset` on `nexus configure`. Use the same flags on `nexus auto`. **0.10.0** October 20, 2026 **New** * **FireConnect is now Nexus CLI.** The command is `nexus`, and your harnesses, key, and settings carry over. * **FireRouter is now Nexus Auto**, with the model ID `auto` in every harness. * **Keep Nexus Auto on Fireworks models** with `nexus auto --providers fireworks`. * **[Claude Desktop](https://fireworks.ai/docs/nexus/harnesses/claude-desktop.md) runs on Linux** as well as macOS. **Deprecated and removed** * Removed FireRouter models and the `auto-instant` mix. * Deprecated `FIRECONNECT_*` environment variables. * Deprecated `--routing-preference`. - Rename: your data moves from `~/.fireconnect` to `~/.nexus`. The old path stays as a link. Try `nexus status`. - The update moves every harness on a FireRouter route to `auto`. - `nexus auto --providers fireworks` is global, and `nexus auto --reset` restores the default. See [Nexus Auto](https://fireworks.ai/docs/nexus/auto.md#which-models-auto-can-use). - Removed models: `firerouter`, `firerouter/...`, and `auto-instant`. Picking one fails and points you to `auto`. - `FIRECONNECT_*` variables still work. Rename them to `NEXUS_*` in your scripts and shell profile. - `--routing-preference` has no effect. Set [effort](https://fireworks.ai/docs/nexus/orchestration-preferences.md) in your harness instead. **0.9.9** October 12, 2026 **New** * **[Codex](https://fireworks.ai/docs/nexus/harnesses/codex.md) starts on FireRouter (now Nexus Auto)**, and OpenAI turns use your own sign-in. * **MCP servers and ChatGPT app connectors work in Codex.** * **Rotating your Fireworks key no longer breaks connected harnesses.** * **OpenCode v2 lists your Fireworks models.** **Fixed** * OpenCode's `/models` no longer lists models twice. * Claude Desktop works behind TLS-inspecting corporate proxies. * Node.js from fnm no longer goes missing in new shells. * When the model catalog can't load, you see the gateway's actual error. **Deprecated and removed** * Removed `--anthropic-api-key` from `nexus claude`. - Codex OpenAI sign-in: run `codex login` with ChatGPT or an OpenAI key. Easy turns stay on open models, and the OpenAI models your sign-in offers sit above the Fireworks catalog in `/model`. - Codex MCP servers and ChatGPT app connectors: their tools now reach the model. - Key rotation: `nexus login` saves the new key into every connected harness, VS Code included, even when you're already signed in. - OpenCode v2: when `opencode.json` uses the v2 `providers` map, Nexus CLI writes Fireworks there too. - OpenCode duplicate models: for example `glm-latest`. - Claude Desktop proxy fix covers proxies such as Zscaler. Run `nexus claude-desktop on` again to pick it up. - fnm fix: the CLI could stop finding Node in new shells. Re-run the installer once to fix it. - Anthropic turns in Claude Code use Claude Code's own sign-in. **0.9.8** October 4, 2026 **New** * **Claude Desktop runs on Fireworks** with `nexus claude-desktop on`. * **Pin Claude Code tiers** to open models, such as for subagents. * **FireRouter (now Nexus Auto) can use OpenAI models** with your own OpenAI key. * **New Anthropic and OpenAI models show real prices.** * **`nexus models`** is a shortcut for `nexus model list`. **Fixed** * `nexus claude --model ` now makes the model your default. * Codex auto-review runs on the model you picked. * `--model` on a first Codex connection keeps the full catalog. * DeepSeek Harness 0.1.7 and later skip the DeepSeek key prompt. If you use Codex auto-review or DeepSeek Harness, run `nexus codex` or `nexus deepseek` once after updating. * Claude Desktop (macOS): sends Chat, Cowork, and Code through Fireworks, starting on `auto`. `off` puts your previous setup back exactly as it was. * Tier pinning keeps Claude as your main model and hands subagents to an open model: `nexus claude --sonnet deepseek-flash-latest --subagent deepseek-flash-latest`. Pass `native` to release a slot. * FireRouter with OpenAI: Cursor IDE, the Copilot harnesses, and DeepSeek Harness can pick it too. * Prices appear as soon as models.dev lists them. Prices come from a cached snapshot, so session estimates still work offline. * `nexus claude --model ` also adds the model to `/model`. * Codex auto-review previously denied every approval. * DeepSeek Harness 0.1.7 and later go straight to Fireworks. **0.9.7** September 21, 2026 **New** * **GitHub Copilot works with Fireworks.** `nexus copilot-app` connects the desktop app, and `nexus copilot-cli` connects the CLI. * **Claude Code keeps its own tier models.** * **Every other harness starts on `auto`.** * **Claude Code gets native [web search](https://fireworks.ai/docs/nexus/auto.md#web-search) and fetch.** * **Codex web search works again.** **Improved** * The built-in Explore agent follows your session model again. * Connecting again refreshes the picker. * Codex compacts long sessions so they can continue. * GLM 5.3 models unlock the max reasoning effort. **Removed** * The `fireworks-websearch` MCP server. - Copilot: your own models stay, and `off` removes only what Nexus CLI added. - Claude Code tiers: the Fireworks catalog joins the `/model` picker instead of replacing slots. - Claude Code web search and fetch are served by Fireworks. - Explore agent previously always used Opus. - Refreshing the picker: retired models go, new ones arrive, and pricing updates. - Codex compacts at 80% of the context window. - The update removes `fireworks-websearch` and leaves your own MCP servers alone. **0.9.6** September 3, 2026 **Fixed** * Codex 0.153 works with Fireworks again. * The status line and session meter name each model once, instead of listing it twice. * Claude Fable 5.1 pricing is in estimates and `nexus model list`. * A routed subagent slot keeps its `[1m]` tag. * `nexus model list` lines up prices across models. * `nexus claude demo` turns off your own MCP servers so they can't skew the race. **0.9.5** August 31, 2026 **New** * **Session cost in Claude Code.** A [status line](https://fireworks.ai/docs/nexus/metrics.md) shows which models ran and what they cost. * **Support for the [ChatGPT desktop app](https://fireworks.ai/docs/nexus/harnesses/chatgpt.md).** * **Pick a model for each Claude Code slot** in the setup wizard. * **Faster `nexus model list`.** **Improved** * MCP tool search works through the Fireworks gateway again. * `nexus claude demo` and the status line use the same prices. * The update repairs missing dependencies. * Better key storage on Linux over SSH or WSL. * `nexus claude status` shows every slot. * GLM 5.3 and GLM 5.3 Flash are in the catalog. - Status line: costs at Fireworks rates plus your cache savings. If you already have a status line, Nexus CLI leaves it alone. - ChatGPT: `nexus chatgpt` and `nexus codex` share one config, so one command connects both. - `nexus model list`: results are cached for an hour, work offline, and show when they were last updated. `--refresh` pulls the latest. - Demo and status line numbers now match. - Repairing dependencies lets a partial install recover. - Linux over SSH or WSL: keys go to an encrypted file, with timeouts on keyring and network calls. - `nexus claude status` includes defaults you never set. **0.9.4** August 20, 2026 **Improved** * `nexus claude demo` estimates cost more accurately. * Connecting and disconnecting Cursor IDE is more reliable. * `nexus uninstall` restores one harness at a time and waits for open IDEs to quit. **0.9.3** August 18, 2026 **New** * **Claude Code can use Anthropic and Fireworks models side by side.** * **`nexus claude demo`** races two live Claude Code sessions side by side. * **[DeepSeek Harness](https://fireworks.ai/docs/nexus/harnesses/deepseek.md) replaces Deep Agents.** Connect it with `nexus deepseek`. **Fixed** * `nexus status` no longer reports "All off" while models are connected. * OpenCode shows the cost of each model. **Deprecated** * `nexus demo`, replaced by `nexus claude demo`. **0.9.2** August 9, 2026 **New** * **VS Code uses the Chat Completions API** for better compatibility. * **Cursor IDE's picker hides the built-in models** while connected. * **A live session cost meter for Claude Code**, now opened with `nexus claude usage`. **Fixed** * Kimi K3 lists its 1M-token context window. * On Linux, quit detection ignores VS Code's Electron helper processes. **0.9.1** August 1, 2026 **New** * **Claude Code stays connected** when you reinstall or update. * **Update in place.** When a newer version is available, interactive terminals offer to update. * **Claude Code model choices persist** across sessions, with a guided setup. **Improved** * More reliable setup for Cursor IDE, OpenCode, and VS Code. * Clearer Claude Code model defaults and catalog labels. - Update prompt: answer No to snooze for a day. - Catalog labels include Kimi K3 naming and pricing. **0.9.0** July 26, 2026 **New** * **`nexus model list`** shows the coding model catalog in one place. * **FireRouter (now Nexus Auto) works like any other model** through `--model`. * **Claude Code usage reports** show session tokens and estimated Fireworks cost. **Improved** * Sign-in supports enterprise SSO and stores your key securely. * Your own settings, headers, and model selections are kept when you connect and disconnect. * More reliable Windows installs and error messages. - `nexus model list` covers standard and fast options across model families. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.