# WidelAI

> WidelAI is a multi-model AI workspace. One account gives access to models from Anthropic, OpenAI, Google, Moonshot AI, and Zhipu AI on the web, in a native macOS coding agent (WidelCode), and on Android, billed in credits at published per-model rates.

- Plans: Starter $19/month (2,000 credits) includes a curated set of cost-efficient models: GPT-4o mini, GPT-3.5 Turbo, Claude Sonnet 4.5, Gemini 3.5 Flash Lite, and GLM-5.3-Flash. Pro $49/month (7,000 credits) includes the full catalog. Enterprise is priced on request.
- Credits: one credit equals $0.005 of provider API cost. Each model has separate input and output rates per 1K tokens, equal to the provider's price per 1M tokens multiplied by 0.2. When credits run out, generation pauses instead of creating an overage charge.
- Privacy: data is encrypted in transit and at rest, and prompts, files, conversations, and outputs are never used to train AI models. WidelAI is not SOC 2 certified.
- Model switching: a conversation can move between models without losing its history.

## Product

- [Pricing](https://www.widelai.com/pricing): Starter $19/month with 2,000 credits, Pro $49/month with 7,000 credits, Teams $25/seat/month with 3,000 pooled credits per seat, plus Enterprise. Every plan includes WidelCode (Mac app, CLI and VS Code extension) and cloud agents.
- [Free Claude month for startups](https://www.widelai.com/claude-startup): WidelAI is passing on $1,000 in Claude credits: approved builders get a month of Starter (2,000 credits) limited to Claude models, claimed with a card at $0 and ending automatically.
- [Pricing transparency](https://www.widelai.com/pricing/transparency): Input and output credit rates per 1K tokens for every model.
- [AI models](https://www.widelai.com/models): Model cards with IDs, release dates, knowledge cutoffs, and capabilities.
- [Models library](https://www.widelai.com/models-library): Every model in the catalog, including deprecated ones, with availability status.
- [Download WidelCode](https://www.widelai.com/download): Native macOS coding agent that edits files and runs commands in a local project, plus install steps for the CLI and VS Code extension.
- [WidelCode CLI on npm](https://www.npmjs.com/package/widelcode): The coding agent for the terminal on macOS, Linux and Windows: `npm install -g widelcode`, then `widelcode login`. Requires Node.js 20+.
- [WidelCode for VS Code](https://marketplace.visualstudio.com/items?itemName=widelai.widelcode): VS Code extension `widelai.widelcode` (VS Code 1.95+); the same ID is on Open VSX for Cursor, VSCodium and Kiro: https://open-vsx.org/extension/widelai/widelcode
- [WidelAI plugin for ChatGPT](https://www.widelai.com/integrations/chatgpt/widelai): The WidelAI plugin brings the models on your WidelAI account into ChatGPT. Ask one model, compare several on the same question, or open Second opinion beside a conversation and add the answer you prefer to the chat.
- [WidelCode plugin for ChatGPT](https://www.widelai.com/integrations/chatgpt/widelcode): The WidelCode plugin starts cloud coding agents on your GitHub repositories from ChatGPT. Describe a change, and an agent edits the code on a new branch, runs your tests, pushes it and can open a pull request while you keep chatting.
- [WidelAI for Android](https://play.google.com/store/apps/details?id=com.widelai.app): Native Android app on Google Play.
- [Sign up](https://www.widelai.com/auth/signup): Create an account; an active plan is required to generate responses.

## Models

Current models in the catalog, with API model IDs for the newest releases. Rates are on the pricing transparency page.

- Anthropic: Claude Opus 5.5 (`claude-opus-5-5`), Claude Sonnet 5.5 (`claude-sonnet-5-5`), Claude Fable 5.1 (`claude-fable-5-1`), Claude Opus 5, Claude Sonnet 5, Claude Opus 4.5 to 4.8, Claude Sonnet 4.5 and 4.6
- OpenAI: GPT-6 Astra (`gpt-6-astra`), GPT-6 Sol (`gpt-6-sol`), GPT-6 Luna (`gpt-6-luna`), GPT-5.6 Sol, Terra and Luna, GPT-5.5, GPT-5.4, GPT-4o mini
- Google: Gemini 3.8 Flash (`gemini-3.8-flash`), Gemini 3.5 Flash, Gemini 3.5 Flash Lite, Gemini 3.1 Pro
- Moonshot AI: Kimi K3.1 Pro (`kimi-k3.1-pro`), Kimi K3 (`kimi-k3`)
- Zhipu AI: GLM-5.3 (`glm-5.3`), GLM-5.3-Flash (`glm-5.3-flash`)

## Blog

- [Kimi K3.1 Pro vs Opus 5.5, GPT-6 Astra, Fable 5.1, DeepSeek & GLM](https://www.widelai.com/blog/kimi-k3-1-pro-vs-opus-5-5-gpt-6-astra-fable-5-1-deepseek-glm): Kimi K3.1 Pro scores 55 on the Artificial Analysis index at a third of Opus 5.5's cost per task. We compare it with Opus 5.5, GPT-6 Astra, Fable 5.1, DeepSeek V4.1 and GLM-5.3.
- [Kimi K3.1 Pro Is Now Available on WidelAI](https://www.widelai.com/blog/kimi-k3-1-pro-now-on-widelai): Kimi K3.1 Pro is live on WidelAI. It scores 65.7% on Terminal-Bench 4.0, cuts hallucinations from 23.4% to 8.3%, and costs a third less than Kimi K3 at 0.40/2.00 credits per 1K tokens.
- [Muse Wonder vs Opus 5.5, Fable 5.1, GPT-6 Astra, Sol and Argon](https://www.widelai.com/blog/muse-wonder-vs-opus-5-5-astra-gemini-4-argon): Meta's Muse Wonder Full ranks #2 on the Artificial Analysis index. We compare Muse Wonder and Full with Opus 5.5, Fable 5.1, GPT-6 Astra, GPT-6.1 Sol and Gemini 4 Argon.
- [The WidelCode CLI and VS Code Extension Are Now Available](https://www.widelai.com/blog/widelcode-cli-vs-code-extension-release): The WidelCode CLI is on npm and the VS Code extension is on the Marketplace and Open VSX. Install steps, sign-in, safety defaults, CI use and current limits.
- [GPT-6.1 Sol Is Now Available in WidelCode](https://www.widelai.com/blog/gpt-6-1-sol-now-in-widelcode): GPT-6.1 Sol is live in WidelCode. It ties GPT-6 Astra on DeepSWE at a fifth of the price, halves cached input to $0.10, and costs 0.40/2.00 credits per 1K tokens.
- [GPT-6.1 Sol vs Claude Sonnet 5.5 vs Opus 5.5: Which to Use?](https://www.widelai.com/blog/gpt-6-1-sol-vs-claude-sonnet-5-5-vs-opus-5-5): GPT-6.1 Sol and Claude Sonnet 5.5 share a $2/$10 price; Opus 5.5 costs double. We compare caching, long-prompt pricing, vendor benchmarks and coding-agent fit.
- [Claude Sonnet 5.5 Is Now Available on WidelAI](https://www.widelai.com/blog/claude-sonnet-5-5-now-on-widelai): Claude Sonnet 5.5 is live on WidelAI. It scores 70.6% on Terminal-Bench 4.0, runs 30%+ faster than Sonnet 5, and costs 0.40/2.00 credits per 1K tokens. Here's what changed.
- [Claude Sonnet 5.5 vs Opus 5.5: Which Claude Should You Use?](https://www.widelai.com/blog/claude-sonnet-5-5-vs-opus-5-5): Sonnet 5.5 costs half as much as Opus 5.5 per token and lands within a few points on Anthropic's benchmarks. Here's when the cheaper Claude wins and when Opus 5.5 earns its rate.
- [How to Use Claude Sonnet 5.5: Effort, Prompts, and Cost](https://www.widelai.com/blog/how-to-use-claude-sonnet-5-5-widelai): A practical Claude Sonnet 5.5 guide: pick the right reasoning effort, write prompts that set scope, get reliable JSON and chart reads, and route work to keep credit costs low.
- [Claude Sonnet 5.5 Migration Guide: 5 Breaking API Changes](https://www.widelai.com/blog/claude-sonnet-5-5-migration-guide): Moving from Claude Sonnet 5 to Sonnet 5.5? Handle five breaking API changes: between_tools thinking, no forced tool use, bound thinking blocks, a new computer-use toolset, and more.
- [AI Coding Agents Compared 2026: Claude Code, Codex, Cursor & WidelCode](https://www.widelai.com/blog/ai-coding-agents-comparison-2026): Claude Code, Codex, Cursor, Kiro, Copilot, Antigravity, Devin Desktop and WidelCode compared for 2026: pricing, Claude Sonnet 5.5 and GPT-6.1 Sol, features, and cost control.
- [New AI Models on WidelAI: Opus 5.5, GPT-6 & Gemini 3.8](https://www.widelai.com/blog/new-ai-models-widelai-september-2026): Meet Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and Gemini 3.8 Flash on WidelAI, with verified specs, credit rates, routing guidance, and rollout considerations.
- [Claude Opus 5.5 vs GPT-6 Sol: Coding, Agents, and Cost](https://www.widelai.com/blog/claude-opus-5-5-vs-gpt-6-sol): Compare Claude Opus 5.5 and GPT-6 Sol across coding, agent design, reasoning controls, context, cache economics, direct prices, and WidelAI credit rates.
- [GPT-6 Sol vs GPT-6 Luna: Which Model Should You Choose?](https://www.widelai.com/blog/gpt-6-sol-vs-gpt-6-luna): Compare GPT-6 Sol and GPT-6 Luna on coding, high-volume work, reasoning, tools, long-context pricing, and a measured 20× cost-aware escalation strategy.
- [Gemini 3.8 Flash Guide: Coding, Agents, Price, and Migration](https://www.widelai.com/blog/gemini-3-8-flash-guide): Learn Gemini 3.8 Flash for coding and agents, including its 1M context, introductory pricing, thinking levels, API migration steps, and safe rollout plan.
- [GPT-6 Astra vs Claude Fable 5.1: Which Model Should You Use?](https://www.widelai.com/blog/gpt-6-astra-vs-claude-fable-5-1): Compare GPT-6 Astra and Claude Fable 5.1 on coding, agents, context, pricing, safeguards, and workflow fit—and learn how to use both in WidelAI and WidelAI Code.
- [How to Use GPT-6 Astra on WidelAI: A Practical Guide](https://www.widelai.com/blog/how-to-use-gpt-6-astra-widelai): Learn how to use GPT-6 Astra on WidelAI: write stronger prompts, select reasoning effort, control long-context costs, design tool workflows, and migrate safely from GPT-5.6 Sol.
- [GPT-6 Astra vs GPT-5.6 Sol: Specs, Price, and Use Cases](https://www.widelai.com/blog/gpt-6-astra-vs-gpt-5-6-sol): Compare GPT-6 Astra and GPT-5.6 Sol on verified specs, token price, reasoning, tool workflows, migration risk, and real use cases—without invented benchmark scores.
- [GPT-6 Astra Is Now Available on WidelAI](https://www.widelai.com/blog/gpt-6-astra-now-on-widelai): GPT-6 Astra is now on WidelAI Pro, with 1.05M context, five reasoning levels, modern tool workflows, and transparent 2/10 credit rates. Here are the verified details.
- [How to Use Claude Fable 5.1 on WidelAI: A Practical Guide](https://www.widelai.com/blog/how-to-use-claude-fable-5-1-widelai): Claude Fable 5.1 is the most capable model on WidelAI and the most expensive. Here's how to prompt it, exploit its 75% cheaper cache reads, and route work between tiers so it earns its rate.
- [Claude Fable 5.1 vs Fable 5 vs Opus 5: Which Anthropic Model Should You Use?](https://www.widelai.com/blog/claude-fable-5-1-vs-fable-5-vs-opus-5): Fable 5.1 leads every benchmark Anthropic published, but Opus 5 costs half as much and lands within a few points on everyday work. A clear breakdown of when each Anthropic flagship is the right call.
- [Claude Fable 5.1 Is Now Available on WidelAI](https://www.widelai.com/blog/claude-fable-5-1-now-on-widelai): Claude Fable 5.1, Anthropic's most advanced model, is now live on WidelAI. It doubles Fable 5's agentic research score at the same token price, with cache reads 75% cheaper. Here's what it does and how to use it.
- [GLM-5.3-Flash (Ox-Alpha) Is Now Available on WidelAI](https://www.widelai.com/blog/glm-5-3-flash-now-on-widelai): GLM-5.3-Flash, the model tested anonymously as "Ox-Alpha," is now live on WidelAI at the lowest credit rate on the platform. Here's what this natively multimodal, million-token model does, how it benchmarks, and how it's priced.
- [WidelAI for Mac Is Here, With a Coding Agent That Runs Locally](https://www.widelai.com/blog/widelai-mac-app-download): WidelAI is now a native Mac app with a coding agent that edits your real files and runs commands in your own shell. Every change arrives as a reviewable diff, every run can be reverted, and it is signed and notarized by Apple.
- [Kimi K3 vs GLM-5.3: Which Open-Weight-Era Frontier Model Should You Use?](https://www.widelai.com/blog/kimi-k3-vs-glm-5-3): Moonshot's Kimi K3 and Zhipu's GLM-5.3 both deliver frontier-class reasoning and a 1M-token context window below Western flagship prices. We compare them on capability, cost, and real-world fit to help you choose.
- [GLM-5.3 Is Now Available on WidelAI](https://www.widelai.com/blog/glm-5-3-now-on-widelai): GLM-5.3, Zhipu AI's coding- and agent-focused flagship with a 1M-token context window, is now live on WidelAI at the lowest credit rate on the platform. Here's what it does, how it's priced in credits, and how to start using it in under a minute.
- [WidelAI for Android Is Now on Google Play](https://www.widelai.com/blog/widelai-android-app-google-play): WidelAI is now a native Android app on Google Play. Switch between OpenAI, Anthropic, and Google models mid-conversation, attach photos and files, track credits per model, and lock it behind your fingerprint.
- [Claude Opus 5 Is Now Available on WidelAI](https://www.widelai.com/blog/claude-opus-5-now-on-widelai): Claude Opus 5, Anthropic's new flagship, is now live on WidelAI. It comes close to Fable 5's capabilities at half the price. Here's what it does, how it's priced in credits, and how to start using it in under a minute.
- [Claude Opus 5 vs Claude Fable 5: Which Anthropic Model Should You Use?](https://www.widelai.com/blog/claude-opus-5-vs-fable-5): Claude Opus 5 comes close to Fable 5's capabilities at half the price. Here's a clear breakdown of how the two Anthropic models compare on capability, cost, and real-world fit — and how to choose.
- [Claude Opus 5 vs GPT-5.6 Sol: The 2026 Frontier Coding Showdown](https://www.widelai.com/blog/claude-opus-5-vs-gpt-5-6-sol): OpenAI's GPT-5.6 Sol and Anthropic's Claude Opus 5 both shipped in July 2026. We compare them on coding, price, context, and tooling to help you decide which frontier workhorse to use.
- [Kimi K3 Is Now Available on WidelAI](https://www.widelai.com/blog/kimi-k3-now-on-widelai): Kimi K3, Moonshot AI's 2.8T-parameter open-weight flagship with a 1M-token context window, is now live on WidelAI. Here's what it does, how it's priced in credits, and how to start using it in under a minute.
- [GPT-5.6 Sol, Terra, and Luna Are Now Available on WidelAI](https://www.widelai.com/blog/gpt-5-6-sol-terra-luna-now-on-widelai): OpenAI's GPT-5.6 family — Sol, Terra, and Luna — is now live on WidelAI. Here's what each tier does, how they're priced in credits, and how to start using them in under a minute.
- [Claude Fable 5 Is Now Available on WidelAI](https://www.widelai.com/blog/claude-fable-5-now-on-widelai): Claude Fable 5, Anthropic's Mythos-class flagship, is back online and live on WidelAI. Here's what it brings, how it's priced in credits, and how to start using it in under a minute.
- [Claude Sonnet 5 Is Now Available on WidelAI](https://www.widelai.com/blog/claude-sonnet-5-now-on-widelai): Claude Sonnet 5, Anthropic's newest model, is now live on WidelAI. Here's what it brings, how it's priced in credits, and how to start using it in under a minute.
- [The Future of Fable 5 and the Rise of GLM-5.2 Open Weights](https://www.widelai.com/blog/future-of-fable-5-rise-of-glm-5-2-open-weights): Anthropic's Fable 5 was pulled days after launch under a US export directive, while Z.ai's GLM-5.2 went fully open under MIT. Here's what both mean for building a resilient AI strategy in 2026.
- [Best Platforms to Market Your AI SaaS in 2026](https://www.widelai.com/blog/best-platforms-market-ai-saas-2026): From LinkedIn to Reddit to Google Ads — a data-backed guide to the platforms that deliver the highest-paying, most genuine users for AI SaaS products in 2026.
- [WidelAI vs Ryne AI: Which Multi-Model AI Platform Delivers More Value?](https://www.widelai.com/blog/widelai-vs-ryne-ai): Ryne AI is an academic humanizer suite with a multi-model chat attached. WidelAI is a frontier-model workspace with a published rate card. Updated with verified 2026 pricing and model lineups.
- [WidelAI vs FiestaAI: A Detailed Platform Comparison for 2026](https://www.widelai.com/blog/widelai-vs-fiesta-ai): How does WidelAI stack up against FiestaAI on model variety, security compliance, credit-based pricing, and developer experience? We break it all down.
- [Getting Started with WidelAI: Your First 30 Minutes](https://www.widelai.com/blog/getting-started-with-ai-models): A practical walkthrough of signing up, choosing a plan, picking your first model, and sending your first prompt on WidelAI.
- [AI Model Landscape in 2026: What Builders Need to Know](https://www.widelai.com/blog/future-of-ai-2026): From GPT-5.2 to Gemini 3.1 Pro to Claude Opus 4.7 — a look at the top OpenAI, Google, and Anthropic models shaping 2026 and how to access them all from one place.
- [Building Production AI Apps with a Multi-Model Platform](https://www.widelai.com/blog/building-production-ai-applications): Architecture patterns, fallback strategies, and cost management tips for teams shipping AI-powered products on WidelAI.

## Optional

- [Full LLM context](https://www.widelai.com/llms-full.txt): Long-form detail on every model, plan, platform, and blog post.
- [Contact](https://www.widelai.com/contact): Sales and support enquiries.
- [Terms of Service](https://www.widelai.com/terms): Billing, refunds, and credit rules.
- [Privacy Policy](https://www.widelai.com/privacy): How account and content data is handled.
- [Cookie Policy](https://www.widelai.com/cookies): Cookies and similar technologies.
- [Accessibility](https://www.widelai.com/accessibility): Accessibility statement.