Claude Code & Codex in one platform
Subscribe to the models you need. Claude Code for Claude, Codex for GPT. One-click CLI setup, monthly credits, OpenAI-compatible API.
Trusted by engineering teams at forward-thinking companies
15M+
API Requests Daily
99.99%
Uptime
5,200+
Developers
<100ms
Avg Latency
Two product lines, one platform
Codex for GPT models. Claude Code for Claude models. Subscribe to one or both — each has independent credits.
Everything you need to ship AI features faster
A complete platform for LLM orchestration — from development to production.
Full Model Access
Every tier unlocks all models in your chosen family — Claude Code gives you Sonnet, Opus, Haiku, Fable. Codex gives you GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.6 Luna.
One-Click CLI Setup
Copy two config files, restart your terminal, and you're done. Claude Code CLI and Codex CLI work out of the box with zero configuration.
Usage Analytics
Real-time dashboard with daily spending charts, trend tracking, and anomaly detection. Know exactly where your credits are going.
Budget Controls
Set custom spending thresholds and get alerted before you exceed them. Hard and soft limits keep your costs predictable.
Global Low Latency
Edge-optimized infrastructure delivers consistent performance worldwide. Streaming responses with real-time token billing.
OpenAI-Compatible API
Drop-in replacement for the OpenAI SDK. Change the base URL, use your API key, and every tool in the ecosystem just works.
Start building in minutes
Pick a plan, copy your config, start coding. It takes less than a minute.
Pick your plan
Choose Claude Code or Codex — or both. Plans from $5/mo with monthly credits. Every tier unlocks all models in that family.
Copy your config
Generate an API key, copy the ready-to-use config files from your dashboard. No manual setup, no environment variables to hunt down.
Start building
Restart your terminal, run claude or codex, and you're coding. Also works with any OpenAI-compatible SDK via the REST API.
Claude Code — CLI Config
Copy these two files, restart terminal, and run claude.
# ~/.claude/settings.json
{
"env": {
"ANTHROPIC_AUTH_TOKEN": "sk-xxx",
"ANTHROPIC_BASE_URL": "https://fastmodel.dev",
"ANTHROPIC_MODEL": "claude-sonnet-5",
"ANTHROPIC_DEFAULT_HAIKU_MODEL": "claude-haiku-4.5",
"ANTHROPIC_DEFAULT_OPUS_MODEL": "claude-opus-4.8",
"ANTHROPIC_DEFAULT_FABLE_MODEL": "claude-fable-5",
"CLAUDE_CODE_EFFORT_LEVEL": "max"
}
}
# ~/.claude.json
{ "hasCompletedOnboarding": true }
Codex — CLI Config
Set your API key, copy the config, and run codex.
# ~/.zshrc or ~/.bashrc export FASTMODEL_API_KEY="sk-xxx" # ~/.codex/config.toml model = "gpt-5.6-luna" model_provider = "fastmodel" model_reasoning_effort = "high" model_supports_reasoning_summaries = true model_reasoning_summary = "none" model_context_window = 131072 model_catalog_json = "model-catalogs/fastmodel-models.json" web_search = "cached" [model_providers.fastmodel] name = "fastmodel" base_url = "https://fastmodel.dev/v1" env_key = "FASTMODEL_API_KEY" wire_api = "responses"
Loved by engineering teams
See why developers choose FastModel for their AI infrastructure.
“The Claude Code CLI setup took 30 seconds. Two config files, restart terminal, and I was coding with Opus. Absolutely seamless.”
Sarah Chen
CTO, DataForge AI
“Switched from managing API keys across multiple accounts to a single subscription. The usage dashboard alone is worth it.”
Marcus Rivera
Lead Engineer, Synthwave Labs
“We subscribe to both Codex and Claude Code. Being able to pick the right model for each task without juggling providers is a game changer.”
Priya Sharma
VP Engineering, Nucleus
“The config preview is brilliant. Copy, paste, done. No hunting through docs for environment variable names or base URLs.”
David Kim
Engineering Manager, ScaleGrid
“Budget alerts saved us from a surprise bill in our first week. Set a threshold, get notified, sleep easy. Simple and effective.”
Alex Thompson
Senior Engineer, TitanML
“Having all Claude models available on every tier is the right call. We started on Basic, upgraded when we needed more credits. Zero friction.”
Jordan Lee
Infrastructure Lead, Meridian AI
“The OpenAI-compatible API means every tool in our stack just works. LangChain, Vercel AI SDK — zero code changes.”
Taylor Wright
Founder, Prism Analytics
“From signup to production in under 10 minutes. FastModel let us focus on our product instead of wrangling API integrations.”
Morgan Wu
Head of AI, Veridian Tech
“The Claude Code CLI setup took 30 seconds. Two config files, restart terminal, and I was coding with Opus. Absolutely seamless.”
Sarah Chen
CTO, DataForge AI
“Switched from managing API keys across multiple accounts to a single subscription. The usage dashboard alone is worth it.”
Marcus Rivera
Lead Engineer, Synthwave Labs
“We subscribe to both Codex and Claude Code. Being able to pick the right model for each task without juggling providers is a game changer.”
Priya Sharma
VP Engineering, Nucleus
“The config preview is brilliant. Copy, paste, done. No hunting through docs for environment variable names or base URLs.”
David Kim
Engineering Manager, ScaleGrid
“Budget alerts saved us from a surprise bill in our first week. Set a threshold, get notified, sleep easy. Simple and effective.”
Alex Thompson
Senior Engineer, TitanML
“Having all Claude models available on every tier is the right call. We started on Basic, upgraded when we needed more credits. Zero friction.”
Jordan Lee
Infrastructure Lead, Meridian AI
“The OpenAI-compatible API means every tool in our stack just works. LangChain, Vercel AI SDK — zero code changes.”
Taylor Wright
Founder, Prism Analytics
“From signup to production in under 10 minutes. FastModel let us focus on our product instead of wrangling API integrations.”
Morgan Wu
Head of AI, Veridian Tech
Frequently asked questions
Everything you need to know about FastModel.
What is FastModel?
FastModel provides managed API access to frontier AI models through two product lines — Claude Code for Anthropic's Claude models, and Codex for OpenAI's GPT models. Pick one or both, copy the config, and start building.
How does pricing work?
Each product line has 4 tiers — Basic ($5), Standard ($20), Pro ($50), and Max ($100). Every tier gives you full access to all models in that family. The only difference is your monthly credit amount. Subscribe to both lines if you need both GPT and Claude.
What's the difference between Codex and Claude Code?
Codex gives you access to OpenAI's GPT models (GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.6 Luna) via the Codex CLI or any OpenAI-compatible SDK. Claude Code gives you access to Anthropic's Claude models (Sonnet, Opus, Haiku, Fable) via the Claude Code CLI or the Anthropic SDK.
Can I subscribe to both Codex and Claude Code?
Yes. Each subscription is independent with its own credits and billing cycle. Many teams subscribe to both so they can use the best model for each task.
Is FastModel compatible with the OpenAI SDK?
Yes — fully. Change the base URL to https://fastmodel.dev/v1 and use your FastModel API key. All OpenAI-compatible tools work out of the box, including LangChain, LlamaIndex, Vercel AI SDK, and more.
Do I need a subscription to use Codex CLI or Claude Code CLI?
Yes. Once you subscribe, your billing page shows ready-to-copy config files with your API key embedded. Copy them to the right locations, restart your terminal, and you're set.
Which models are available on each plan?
Every tier unlocks all models in that family — no restrictions. Codex Basic gets the same GPT models as Codex Max. Claude Code Basic gets the same Claude models as Claude Code Max. The only difference is your monthly credits.
Can I set spending limits?
Yes. You can configure spending thresholds from your dashboard. When you hit a limit, requests are blocked until your next billing cycle or until you top up — no surprise bills.
Do you support prompt caching?
Yes. Both Claude and GPT models support prompt caching, which reduces input costs significantly for repeated prefixes. Reuse the same system prompt or long context across requests to benefit automatically.
Do you log or store our API requests?
We log request metadata (model, tokens, latency) for your usage dashboard, but never request or response bodies. All data is encrypted at rest and in transit.
Ready to start building?
Pick Claude Code or Codex — plans from $5/mo with monthly credits. One-click config, start coding in minutes.