Frequently Asked Questions
Quick answers to the questions we hear most often.
Account
How do I reset my password?
Click 'Forgot Password' on the login page and enter the email address associated with your account. You will receive a password reset link within a few minutes — check your spam folder if it does not arrive. The link expires after 24 hours. If you signed up with a social provider (Google, GitHub, etc.) you do not have a password; simply log in with that provider again.
Can I use social login?
Yes. GabForge supports single sign-on via Google, GitHub, GitLab, Microsoft, and Discord. Click the relevant provider button on the login page. On first sign-in, an account is automatically created using your provider's email address. You can link additional social providers or add a password from your Account Settings page at any time.
Api
How do I get an API key?
Log in to your GabForge account, navigate to Dashboard > API Keys, then click 'Create new key'. Give the key a descriptive name (e.g. 'Local dev', 'CI pipeline'), choose an optional credit limit, and click Create. Your key is shown once — copy and store it securely. You can generate multiple keys and revoke any of them at any time from the same page.
What is the base URL for the API?
The base URL is https://ai.gabforge.ai/v1. The API is fully OpenAI-compatible, so you can point any OpenAI SDK or tool at this base URL and use your GabForge API key as the bearer token. For example, with the Python SDK: openai.base_url = 'https://ai.gabforge.ai/v1' and openai.api_key = 'your-key'. Endpoints include /chat/completions, /completions, /embeddings, and /models.
What models are available via the API?
The primary model is gabforge-coder (32B parameters, optimised for code generation, review, and explanation). A smaller gabforge-coder-7b is available for low-latency tasks. When your local GPU is unavailable or you exceed your local compute budget, claude-sonnet is available as a cloud fallback. Call GET /v1/models to see the current list of available models and their context window sizes.
Billing
What payment methods do you accept?
We accept all major credit and debit cards (Visa, Mastercard, American Express, Discover) processed securely through Stripe. We do not store card details — Stripe handles all payment data in a PCI-DSS Level 1 certified environment. Business customers on the Team or Enterprise plans may request invoice billing. Cryptocurrency payments are not currently supported.
How do credits work?
Credits are consumed per API request based on token usage. One credit roughly corresponds to one thousand tokens processed (input + output combined). Each plan includes a monthly credit allocation that resets on your billing anniversary date. You can purchase top-up credit bundles at any time from Dashboard > Billing. Credits never expire as long as your account is active. Unused monthly plan credits do not roll over to the next month, but purchased top-up bundles do.
General
What is GabForge?
GabForge is an open-source AI development platform that bundles everything a developer needs into a single integrated suite: a VS Code-forked IDE with built-in multi-agent AI (GabForge AI Studio), a custom Linux distribution optimised for AI workloads (GabForge OS), an all-in-one creative suite (GabForge Creative Suite), and a fine-tuned coding API powered by Qwen2.5-Coder models running locally or in the cloud. GabForge is built for the developer who wants AI that works — no cloud lock-in, no usage surprises.
Is GabForge free?
Yes. GabForge offers a generous free tier with no time limit. The IDE, OS, and Creative Suite are fully open-source under the Apache 2.0 / MIT dual licence. The hosted API includes a free allocation of credits each month so you can build and experiment at no cost. Pro cloud features — expanded credits, priority model access, and team collaboration — are available on paid plans. See our Pricing page for details.
What AI models does GabForge use?
GabForge primarily uses Qwen2.5-Coder models fine-tuned on high-quality coding data. The flagship model is gabforge-coder-32b, a 32-billion-parameter model that runs entirely on your local GPU via the built-in llama.cpp inference engine (GGUF Q4_K_M quantisation). When cloud headroom is needed, Claude Sonnet acts as a capable fallback. All models are exposed through an OpenAI-compatible API at https://ai.gabforge.ai/v1 so your existing tooling works without modification.
Didn't find your answer?
Open a ticket and our team will help you out.