# CompanyFabric · One API for AI

> One key, one balance, every major text and code model.
> OpenAI-compatible: base URL https://api.companyfabric.com/v1
> Raw markdown mirrors of every page: append the path to /llm (e.g. /llm/claude-sonnet-4-5-api)

## Pages

- [/models](https://companyfabric.com/models): The full model library, every model behind one API key, with exact prices.
- [/claude-sonnet-4-5-api](https://companyfabric.com/claude-sonnet-4-5-api): Claude Sonnet 4.5 API, pricing, playground, code samples. Frontier-quality text and reasoning through one OpenAI-compatible endpoint.
- [/claude-haiku-4-5-api](https://companyfabric.com/claude-haiku-4-5-api): Claude Haiku 4.5 API, pricing, playground, code samples. Fast, cheap frontier-family model for high-volume tasks.
- [/claude-opus-4-6-api](https://companyfabric.com/claude-opus-4-6-api): Claude Opus 4.6 API, pricing, playground, code samples. The frontier tier for the hardest reasoning and agentic work.
- [/gpt-5-5-api](https://companyfabric.com/gpt-5-5-api): GPT-5.5 API, pricing, playground, code samples. OpenAI's flagship, one key away.
- [/kimi-k3-api](https://companyfabric.com/kimi-k3-api): Kimi K3 API, pricing, playground, code samples. Agentic coding model with strong multi-step tool use.
- [/gemini-3-7-flash-api](https://companyfabric.com/gemini-3-7-flash-api): Gemini 3.7 Flash API, pricing, playground, code samples. Google's current fast model, million-token context.
- [/gemini-3-5-flash-api](https://companyfabric.com/gemini-3-5-flash-api): Gemini 3.5 Flash API, pricing, playground, code samples. The previous stable Flash, still current for most work.
- [/gemini-2-5-pro-api](https://companyfabric.com/gemini-2-5-pro-api): Gemini 2.5 Pro API, pricing, playground, code samples. Google's stable Pro tier, the non-preview reasoning model.
- [/gemini-2-5-flash-api](https://companyfabric.com/gemini-2-5-flash-api): Gemini 2.5 Flash API, pricing, playground, code samples. Long-context workhorse on the stable 2.5 line.
- [/gemini-3-1-pro-preview-api](https://companyfabric.com/gemini-3-1-pro-preview-api): Gemini 3.1 Pro (preview) API, pricing, playground, code samples. Google's newest Pro tier, shipped as a preview.
- [/gemini-3-flash-preview-api](https://companyfabric.com/gemini-3-flash-preview-api): Gemini 3 Flash (preview) API, pricing, playground, code samples. The Gemini 3 Flash preview line.
- [/claude-fable-5-api](https://companyfabric.com/claude-fable-5-api): Claude Fable 5 API, pricing, playground, code samples. Anthropic's first Mythos-class model, the new flagship tier above Opus, tuned for the hardest reasoning and agentic work.
- [/claude-opus-5-api](https://companyfabric.com/claude-opus-5-api): Claude Opus 5 API, pricing, playground, code samples. The Claude 5 generation's deep-work model, maximum-capability reasoning for research, architecture, and hard agentic tasks.
- [/deepseek-v4-pro-api](https://companyfabric.com/deepseek-v4-pro-api): DeepSeek V4 Pro API, pricing, playground, code samples. The heavyweight of the V4 family, deeper reasoning than base V4 while keeping open-weight commodity pricing.
- [/deepseek-v4-flash-api](https://companyfabric.com/deepseek-v4-flash-api): DeepSeek V4 Flash API, pricing, playground, code samples. The V4 family's speed tier, near-V4 quality on everyday calls at the lowest price in the family.
- [/grok-4-6-api](https://companyfabric.com/grok-4-6-api): Grok 4.6 API, pricing, playground, code samples. xAI's flagship, half-million-token context.
- [/grok-4-5-api](https://companyfabric.com/grok-4-5-api): Grok 4.5 API, pricing, playground, code samples. The previous flagship, same rate card.
- [/grok-4-3-api](https://companyfabric.com/grok-4-3-api): Grok 4.3 API, pricing, playground, code samples. Million-token context at roughly half the flagship rate.
- [/glm-5-2-api](https://companyfabric.com/glm-5-2-api): GLM-5.2 API, pricing, playground, code samples. Z.ai's open-weight frontier challenger, strong coding and reasoning with one of the best free tiers in class.
- [/kimi-k2-6-api](https://companyfabric.com/kimi-k2-6-api): Kimi K2.6 API, pricing, playground, code samples. Moonshot's value tier, K3-style agentic tool-calling at half the price, a favorite for high-volume agent loops.
- [/models/anthropic/claude-sonnet-4-5](https://companyfabric.com/models/anthropic/claude-sonnet-4-5): Claude Sonnet 4.5 playground page, try it in the browser, exact price per run.
- [/models/anthropic/claude-haiku-4-5](https://companyfabric.com/models/anthropic/claude-haiku-4-5): Claude Haiku 4.5 playground page, try it in the browser, exact price per run.
- [/models/anthropic/claude-opus-4-6](https://companyfabric.com/models/anthropic/claude-opus-4-6): Claude Opus 4.6 playground page, try it in the browser, exact price per run.
- [/models/openai/gpt-5-5](https://companyfabric.com/models/openai/gpt-5-5): GPT-5.5 playground page, try it in the browser, exact price per run.
- [/models/moonshotai/kimi-k3](https://companyfabric.com/models/moonshotai/kimi-k3): Kimi K3 playground page, try it in the browser, exact price per run.
- [/models/google/gemini-3-7-flash](https://companyfabric.com/models/google/gemini-3-7-flash): Gemini 3.7 Flash playground page, try it in the browser, exact price per run.
- [/models/google/gemini-3-5-flash](https://companyfabric.com/models/google/gemini-3-5-flash): Gemini 3.5 Flash playground page, try it in the browser, exact price per run.
- [/models/google/gemini-2-5-pro](https://companyfabric.com/models/google/gemini-2-5-pro): Gemini 2.5 Pro playground page, try it in the browser, exact price per run.
- [/models/google/gemini-2-5-flash](https://companyfabric.com/models/google/gemini-2-5-flash): Gemini 2.5 Flash playground page, try it in the browser, exact price per run.
- [/models/google/gemini-3-1-pro-preview](https://companyfabric.com/models/google/gemini-3-1-pro-preview): Gemini 3.1 Pro (preview) playground page, try it in the browser, exact price per run.
- [/models/google/gemini-3-flash-preview](https://companyfabric.com/models/google/gemini-3-flash-preview): Gemini 3 Flash (preview) playground page, try it in the browser, exact price per run.
- [/models/anthropic/claude-fable-5](https://companyfabric.com/models/anthropic/claude-fable-5): Claude Fable 5 playground page, try it in the browser, exact price per run.
- [/models/anthropic/claude-opus-5](https://companyfabric.com/models/anthropic/claude-opus-5): Claude Opus 5 playground page, try it in the browser, exact price per run.
- [/models/deepseek/deepseek-v4-pro](https://companyfabric.com/models/deepseek/deepseek-v4-pro): DeepSeek V4 Pro playground page, try it in the browser, exact price per run.
- [/models/deepseek/deepseek-v4-flash](https://companyfabric.com/models/deepseek/deepseek-v4-flash): DeepSeek V4 Flash playground page, try it in the browser, exact price per run.
- [/models/xai/grok-4-6](https://companyfabric.com/models/xai/grok-4-6): Grok 4.6 playground page, try it in the browser, exact price per run.
- [/models/xai/grok-4-5](https://companyfabric.com/models/xai/grok-4-5): Grok 4.5 playground page, try it in the browser, exact price per run.
- [/models/xai/grok-4-3](https://companyfabric.com/models/xai/grok-4-3): Grok 4.3 playground page, try it in the browser, exact price per run.
- [/models/z-ai/glm-5-2](https://companyfabric.com/models/z-ai/glm-5-2): GLM-5.2 playground page, try it in the browser, exact price per run.
- [/models/moonshotai/kimi-k2-6](https://companyfabric.com/models/moonshotai/kimi-k2-6): Kimi K2.6 playground page, try it in the browser, exact price per run.
- [/models/anthropic](https://companyfabric.com/models/anthropic): All anthropic models available through the CompanyFabric API.
- [/models/openai](https://companyfabric.com/models/openai): All openai models available through the CompanyFabric API.
- [/models/moonshotai](https://companyfabric.com/models/moonshotai): All moonshotai models available through the CompanyFabric API.
- [/models/google](https://companyfabric.com/models/google): All google models available through the CompanyFabric API.
- [/models/deepseek](https://companyfabric.com/models/deepseek): All deepseek models available through the CompanyFabric API.
- [/models/xai](https://companyfabric.com/models/xai): All xai models available through the CompanyFabric API.
- [/models/z-ai](https://companyfabric.com/models/z-ai): All z-ai models available through the CompanyFabric API.
- [/models/best-for-coding](https://companyfabric.com/models/best-for-coding): The models developers actually route code through, ranked for agentic coding, tool use, and cost per task, all behind one API key.
- [/models/best-for-reasoning](https://companyfabric.com/models/best-for-reasoning): The models to reach for when the task is hard thinking: planning, math, multi-step analysis, compared on capability and exact cost.
- [/models/best-for-writing](https://companyfabric.com/models/best-for-writing): The models that produce copy a human would ship, long-form drafts, marketing voice, and rewriting, with per-call prices.
- [/models/cheapest](https://companyfabric.com/models/cheapest): The lowest-cost models in the library that are still worth calling, ranked by real price per typical task, not just headline token rates.
- [/models/open-source](https://companyfabric.com/models/open-source): The open-weights models worth running through an API, frontier-adjacent quality with no proprietary lock-in, at commodity prices.
- [/models/free](https://companyfabric.com/models/free): The honest map of free AI model access: starter credits on every account, generous provider free tiers, and open-weight models you can self-host.
- [/compare](https://companyfabric.com/compare): How CompanyFabric compares to other AI gateways.
- [/compare/vs-openrouter](https://companyfabric.com/compare/vs-openrouter): Both route the major models behind one OpenAI-compatible key. The differences: BYOK fees, and whether agents are first-class endpoints.
- [/compare/vs-litellm](https://companyfabric.com/compare/vs-litellm): LiteLLM is a self-hosted proxy you operate; CompanyFabric is a hosted gateway with billing built in. The real comparison is ops burden vs control.
- [/compare/vs-llmgateway](https://companyfabric.com/compare/vs-llmgateway): Both are OpenAI-compatible gateways. The split: open-core self-hosting vs a hosted platform with exact metering and agent endpoints.
- [/compare/vs-portkey](https://companyfabric.com/compare/vs-portkey): Portkey is an AI ops control plane, observability, guardrails, routing policies. CompanyFabric is a metered gateway with billing doors. Different centers of gravity.
- [/alternatives/coderabbit](https://companyfabric.com/alternatives/coderabbit): CodeRabbit charges per seat for AI code review. The alternative: a PR review agent you own, running any model through one metered API key.
- [/alternatives/viktor](https://companyfabric.com/alternatives/viktor): Viktor sells AI employees in Slack at enterprise pricing. The open alternative: agents you own, on open templates, metered per call through one key.
- [/alternatives/openrouter](https://companyfabric.com/alternatives/openrouter): Looking for an OpenRouter alternative? CompanyFabric routes text and code behind one key, with BYOK at 0% platform fee.
- [/integrations](https://companyfabric.com/integrations): Use any model in Claude Code, Cursor, Cline, Continue, Aider, and more.
- [/integrations/claude-code](https://companyfabric.com/integrations/claude-code): Point Claude Code at CompanyFabric and run Claude, GPT-5, Gemini, or open models through one key, same agent, any brain.
- [/integrations/cursor](https://companyfabric.com/integrations/cursor): Add CompanyFabric as an OpenAI-compatible endpoint in Cursor and switch between frontier and open models without changing subscriptions.
- [/integrations/cline](https://companyfabric.com/integrations/cline): Cline supports OpenAI-compatible providers natively, one CompanyFabric key gives the agent every model in the library.
- [/integrations/continue](https://companyfabric.com/integrations/continue): Wire Continue's config to CompanyFabric once and every model in the library is available in your editor.
- [/integrations/aider](https://companyfabric.com/integrations/aider): Aider works with any OpenAI-compatible endpoint, run it on frontier or open models through one CompanyFabric key.
- [/integrations/openclaw](https://companyfabric.com/integrations/openclaw): Point OpenClaw's gateway at CompanyFabric and every hosted agent picks from the full model library with per-agent metering.
- [/use-gemini-with-claude-code](https://companyfabric.com/use-gemini-with-claude-code): Use Gemini with Claude Code through one CompanyFabric key, setup, price, and FAQ.
- [/use-deepseek-with-claude-code](https://companyfabric.com/use-deepseek-with-claude-code): Use DeepSeek with Claude Code through one CompanyFabric key, setup, price, and FAQ.
- [/use-kimi-with-claude-code](https://companyfabric.com/use-kimi-with-claude-code): Use Kimi with Claude Code through one CompanyFabric key, setup, price, and FAQ.
- [/use-gpt-5-5-with-claude-code](https://companyfabric.com/use-gpt-5-5-with-claude-code): Use GPT-5.5 with Claude Code through one CompanyFabric key, setup, price, and FAQ.
- [/use-claude-with-cursor](https://companyfabric.com/use-claude-with-cursor): Use Claude with Cursor through one CompanyFabric key, setup, price, and FAQ.
- [/use-gemini-with-cursor](https://companyfabric.com/use-gemini-with-cursor): Use Gemini with Cursor through one CompanyFabric key, setup, price, and FAQ.
- [/use-deepseek-with-cursor](https://companyfabric.com/use-deepseek-with-cursor): Use DeepSeek with Cursor through one CompanyFabric key, setup, price, and FAQ.
- [/use-claude-with-cline](https://companyfabric.com/use-claude-with-cline): Use Claude with Cline through one CompanyFabric key, setup, price, and FAQ.
- [/use-kimi-with-cline](https://companyfabric.com/use-kimi-with-cline): Use Kimi with Cline through one CompanyFabric key, setup, price, and FAQ.
- [/use-gemini-with-cline](https://companyfabric.com/use-gemini-with-cline): Use Gemini with Cline through one CompanyFabric key, setup, price, and FAQ.
- [/use-claude-with-continue](https://companyfabric.com/use-claude-with-continue): Use Claude with Continue through one CompanyFabric key, setup, price, and FAQ.
- [/use-deepseek-with-continue](https://companyfabric.com/use-deepseek-with-continue): Use DeepSeek with Continue through one CompanyFabric key, setup, price, and FAQ.
- [/use-claude-with-aider](https://companyfabric.com/use-claude-with-aider): Use Claude with Aider through one CompanyFabric key, setup, price, and FAQ.
- [/use-deepseek-with-aider](https://companyfabric.com/use-deepseek-with-aider): Use DeepSeek with Aider through one CompanyFabric key, setup, price, and FAQ.
- [/use-claude-with-openclaw](https://companyfabric.com/use-claude-with-openclaw): Use Claude with OpenClaw through one CompanyFabric key, setup, price, and FAQ.
- [/use-gemini-with-openclaw](https://companyfabric.com/use-gemini-with-openclaw): Use Gemini with OpenClaw through one CompanyFabric key, setup, price, and FAQ.
- [/use-kimi-with-openclaw](https://companyfabric.com/use-kimi-with-openclaw): Use Kimi with OpenClaw through one CompanyFabric key, setup, price, and FAQ.
- [/docs](https://companyfabric.com/docs): CompanyFabric API documentation, endpoints, auth, jobs, errors, BYOK.
- [/docs/quickstart](https://companyfabric.com/docs/quickstart): Get a key, point any OpenAI SDK at the CompanyFabric base URL, and make your first metered call.
- [/docs/authentication](https://companyfabric.com/docs/authentication): Bearer authentication with sk-cf- keys: creation, budget caps, revocation, and how metering attributes spend per key.
- [/docs/chat-completions](https://companyfabric.com/docs/chat-completions): POST /v1/chat/completions, the OpenAI wire format across every text and code model, with streaming and exact usage.
- [/docs/models](https://companyfabric.com/docs/models): How model ids work: canonical 2-segment ids, branded aliases, and the companyfabric/* meta-models, plus the /v1/models listing endpoint.
- [/docs/jobs](https://companyfabric.com/docs/jobs): POST /v1/jobs, the async API for video, image, and audio models: submit with the model in the body, poll until completed, read outputs and exact cost.
- [/docs/errors](https://companyfabric.com/docs/errors): Every error is structured JSON with a code, a human message, and a machine-actionable resolution, built for agents that fix their own failures.
- [/docs/byok](https://companyfabric.com/docs/byok): Route through CompanyFabric on your own provider keys at 0% fee, the vault stores them encrypted and never returns more than the last four characters.
- [/run/deepseek-v4-pro-locally](https://companyfabric.com/run/deepseek-v4-pro-locally): Run deepseek-v4-pro locally, VRAM requirements by quantization, GGUF sizes, run commands, and the metered API alternative.
- [/run/deepseek-v4-flash-locally](https://companyfabric.com/run/deepseek-v4-flash-locally): Run deepseek-v4-flash locally, VRAM requirements by quantization, GGUF sizes, run commands, and the metered API alternative.
- [/run/kimi-k3-locally](https://companyfabric.com/run/kimi-k3-locally): Run kimi-k3 locally, VRAM requirements by quantization, GGUF sizes, run commands, and the metered API alternative.
- [/run/kimi-k2-6-locally](https://companyfabric.com/run/kimi-k2-6-locally): Run kimi-k2-6 locally, VRAM requirements by quantization, GGUF sizes, run commands, and the metered API alternative.
- [/run/glm-5-2-locally](https://companyfabric.com/run/glm-5-2-locally): Run glm-5-2 locally, VRAM requirements by quantization, GGUF sizes, run commands, and the metered API alternative.