Wally

What it does

Wally takes a prompt or a coding task and returns a model reply, and it can start a coding agent on a chosen model. The CLI connects tools such as OpenCode, Claude Code, Codex and Hermes to either a local model or a hosted model from Wally Cloud. Hosted requests go to an OpenAI-compatible chat-completions endpoint, where you send messages and read the reply as a stream.

People use it in two main ways: running a coding agent from the terminal after a one-time browser sign-in, or pointing an existing OpenAI client, SDK or curl command at Wally's base URL and changing the model ID. Wally runs hosted models with zero data retention by default, while local models keep data on the machine. Hosted usage is billed against prepaid credits.

It suits developers and teams who want to use open models through an OpenAI-style interface or a coding agent without managing their own inference. Access requires installing the CLI or creating a Cloud key in the console, plus prepaid credits for hosted runs. The API covers chat models; the pricing table lists per-million-token rates that vary by model, and the console is the place to check current rates.

What it can do

• Run a coding agent (OpenCode, Claude Code, Codex, Hermes, DeepSeek Harness, Prime Agent) on a hosted or local model with one command
• Call hosted open models through an OpenAI-compatible chat-completions API with streaming
• Switch models by changing the model ID string
• Sign in once through the browser with Google or GitHub, without copying an API key for CLI use
• Check spend with wally account usage against prepaid credits
• Run a model locally on your machine with the same CLI commands

Works with

• OpenCode
• Claude Code
• Claude Desktop
• Codex
• Command Code
• Hermes
• OpenClaw
• DeepSeek Harness
• Prime Agent
• Dify

Before you choose

• Hosted usage bills against prepaid credits per million tokens; check current model rates on the pricing page before committing.
• Prompts sent to hosted models go to Wally Cloud under zero data retention by default, but the console playground keeps recent conversations.
• The API is OpenAI-compatible for chat models; you need a Cloud key from the console for direct calls, not for wally account login.

Listed under

• Coding

Get started

Create an account, then install the CLI or create a Cloud key.

1. Install the wally CLI with the install command for your OS from the product page.
2. Run wally account login and sign in with Google or GitHub in the browser.
3. Start a coding agent on a model, for example wally opencode --cloud -m glm-5.3-flash, after buying prepaid credits in the console for hosted runs.

Usage is spent from prepaid credits bought in the console. No free tier or trial is stated on the pricing page. Rates are per million tokens, with cached input priced separately for some models.

Source: runwally.com/pricing, checked 30 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 30 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Wally: frontier open models, one API key ↗runwally.com
• Wally documentation - RunAnywhere Docs ↗docs.runanywhere.ai
• Pricing | Wally ↗runwally.com

Company website

runwally.com: product, pricing and policy pages

Hacker News

Launch posts and discussion

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

manual_review

Last checked 30 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.