selan.ai Gateway Reactor

Compare · nexos.ai and Selan

Is Selan a nexos.ai alternative for Claude Code and Codex?

Only for a team's coding tools, and Selan connects nexos.ai as one of its providers, so a team can run both. nexos.ai is an AI platform for a whole company: chat and agents for business teams, and a gateway API to 200+ models, paid for in credits. Selan runs a team's Claude Code, Codex and Claude Desktop on the Claude and ChatGPT subscriptions you already pay for. If you want one AI platform for everyone, nexos.ai is the better answer.

subscription nexos.ai

How do nexos.ai and Selan compare?

nexos.aiSelan
What it isAn AI platform: Chat and Agents for business teams, and a gateway API to 200+ modelsA hosted gateway, with a CLI that launches Claude Code, Codex and OpenCode through it
What it runs onnexos.ai credits; the gateway can also use your own provider API keysYour Claude, ChatGPT and Ollama Cloud subscriptions first, then your API keys, nexos.ai's among them
A developer's credentialA nexos.ai API key in ANTHROPIC_AUTH_TOKEN, set in the shell profileOne sign-in with Google or Microsoft through selan login; no key on the laptop
ClientsAny app on its API; guides for Claude Code, Codex CLI, Claude Cowork, OpenCode, Roo Code and other coding toolsClaude Code, Codex, OpenCode and Claude Desktop
LimitsCredit limits per member and budget limits per team, set by the ownerSpend limits per person, by day, week or month in UTC, on pay-per-use keys
Request logsEvery API call with model, cost and duration, kept for 7 daysEvery request under the person who made it, on the owner's Usage page
How you pay itThe gateway is usage-based: API credits are prepaid on a paid plan, 1 credit per euro. Chat and Agents are on an Enterprise planA flat monthly plan sized by seats, free for up to three; no share of tokens, since your provider bills you directly

nexos.ai gives a whole company one AI platform. Selan spends the coding subscriptions you already own, and can use your nexos.ai key too.

When is nexos.ai the better answer?

When you want one AI platform for the whole company, not only for developers. nexos.ai gives business teams chat, no-code agents and integrations with Slack, Jira, Google Drive and other tools, and gives engineers a gateway API.

Its gateway covers more than coding agents. It answers in the OpenAI format and in Anthropic's Messages format, and it serves embeddings, images and audio as well as chat. It publishes setup guides for OpenCode, Roo Code, Kilo Code and other coding tools; of those, Selan launches OpenCode.

nexos.ai also offers an EU-hosted gateway that runs Claude on Vertex AI in the EU and GPT on Azure in the EU. Through Selan, no route is pinned to the EU today: Selan connects Claude on Vertex on its Global location. See Claude Code and Codex with EU data residency.

When is Selan the better answer?

When your developers already have Claude or ChatGPT subscriptions. Selan sends each request to a subscription first and falls through to a key billed per token only when the seats are capped.

nexos.ai's documentation describes its own credits and provider API keys, and no way to run on a Claude or ChatGPT subscription. Its Claude Cowork guide pitches the gateway as a way to stop expensing separate Anthropic subscriptions. Through nexos.ai alone, every Claude Code session is billed per token, from credits or a provider key.

nexos.ai's Claude Code setup puts the API key in each developer's shell profile, blanks ANTHROPIC_API_KEY and asks them to /logout of any Anthropic login. selan claude sets the base URL and token for the session it launches. The trade-off is that requests travel through Selan, so if Selan is unavailable those requests fail.

How does Selan use nexos.ai?

An owner pastes the company's nexos.ai API key in Selan. Selan then reads the models that key can reach from nexos.ai's own catalog, so nothing is added to Selan first. A developer names one with the provider before the first colon:

selan claude --model nexos:"Claude Sonnet 4.6"

nexos.ai names its models for people, with spaces, so quote the name. Capitals, dashes, dots and spaces do not change which model you get, so nexos:claude-sonnet-4-6 is the same model.

Claude Code asks every model for adaptive thinking. nexos.ai's own Cowork guide shows a Claude 4.5 model refusing it with a 400 and says to pick a newer model. For Claude 4 models on nexos.ai, Selan rewrites the request into a thinking budget instead.

nexos.ai bills those requests to your credits. Selan records the cost nexos.ai reports on each response under the person who made it, and a person's spend limit counts it. Codex reaches nexos.ai's Claude models too, because Selan translates between the two formats.

The routes are in Providers and models, and naming a model in Run any model.

Good to know: nexos.ai's own guide says Claude Code is optimized for Anthropic models and may not behave correctly with other providers. Anthropic doesn't support routing Claude Code to non-Claude models through any gateway, Selan included.

How do you move Claude Code from nexos.ai to Selan?

nexos.ai's setup exports ANTHROPIC_BASE_URL=https://api.nexos.ai, the key in ANTHROPIC_AUTH_TOKEN, and a model per role such as ANTHROPIC_DEFAULT_SONNET_MODEL. selan claude sets the base URL and token for the session it launches, in place of whatever the shell exported. Delete nexos.ai's exports from the shell profile, the model variables included, because a model name left there reaches Selan without the nexos: before it.

Connect the same nexos.ai key in Selan and its models stay one flag away. Limits per person are on the spend page.

Start for free

Free for up to three seats, and the Free plan needs no card.

The short version

What nexos.ai does is as its home, AI gateway, EU gateway and pricing pages and its Claude Code, Codex CLI, Claude Cowork, API keys, API credits, credit management and observability documentation at docs.nexos.ai described it, read on 23 October 2026. What Selan does is selan-cli, selan-control and selan-proxy as they stood on the same day.