Knowledge Base Sections ▾
Navigation
▸ Start here By rolesCategories
- Cursor + Gonka AI - cheap LLM for coding
- Claude Code + Gonka AI - LLM for the terminal
- OpenClaw + Gonka AI - affordable AI agents
- OpenCode + Gonka AI - free AI for code
- Continue.dev + Gonka AI - AI for VS Code/JetBrains
- Cline + Gonka AI - AI agent in VS Code
- Aider + Gonka AI - pair programming with AI
- LangChain + Gonka AI - AI applications for pennies
- n8n + Gonka AI - automation with cheap AI
- Open WebUI + Gonka AI - your own ChatGPT
- LibreChat + Gonka AI — open-source ChatGPT
- Hermes Agent + Gonka AI — Autonomous Agent for Pennies
- Kilo Code + Gonka AI — AI-Agent in VS Code
- Roo Code + Gonka AI — Autonomous AI Agent in VS Code
- LlamaIndex + Gonka AI — RAG applications for pennies
- PydanticAI + Gonka — typed AI agents for pennies
- Vercel AI SDK + Gonka AI — AI applications in TypeScript for pennies
- TanStack AI + Gonka — AI applications in TypeScript for pennies
- API quick start — curl, Python, TypeScript
- JoinGonka Gateway — a full overview
- Management Keys — SaaS on Gonka
- Cheapest AI API: Provider Comparison 2026
- Cursor Pro request limit reached — breakdown and cheaper alternative
- Claude Code is cheaper — bill breakdown and switching
- Cline is burning money — why the agent spends so much
- OpenClaw is expensive — why the agent burns through tokens and how to save
- OpenRouter: Cheap Alternative — Comparison with JoinGonka Gateway
- Best AI model for coding in 2026: comparison and prices
- Cheap alternative to GitHub Copilot without limits
- A cheap Windsurf alternative without credits or limits
- The cheapest API for AI agents in 2026
- ZCode: Cheap GLM inference instead of GLM Coding Plan
- JetBrains IDE + JoinGonka Gateway — your own endpoint instead of credits
- GitHub Copilot BYOK — own models instead of quotas
- Zed + JoinGonka Gateway — cheap inference in your editor
- Pi + JoinGonka Gateway — terminal agent on cheap inference
copilot-byok.tag
GitHub Copilot BYOK — own models instead of quotas
BYOK (bring your own key) is the official GitHub Copilot mechanism that allows you to connect your own model provider and run agentic sessions on it, rather than on the built-in GPT or Claude models. Since January 2026, any OpenAI-compatible endpoint is supported. As of June 23, BYOK is available in the Copilot app itself in public preview on all paid plans, and as of July 14, in the JetBrains version as well. This means you can keep Copilot as the shell and route inference through the JoinGonka Gateway — at decentralized Gonka network prices and without consuming your subscription's premium request quota.
What BYOK offers and when you need it
A Copilot subscription restricts premium requests with monthly limits; agentic sessions, refactoring, and multi-step tasks consume this quota fastest. BYOK shifts these calls to your provider: consumption is billed at your provider's rates and does not deduct from the Copilot premium quota. Keys are stored in the system keychain and cannot be read back by the interface.
BYOK is justified if you are hitting limits, want models with open weights, or want to pay significantly less for inference: the Gonka network offers the same open-source models hundreds of times cheaper than vendor retail prices — current rates are always available in the live gateway price list.
Timeline: what was included and when
| When | What appeared |
|---|---|
| November 2025 | BYOK in public preview: Anthropic, OpenAI, xAI, Microsoft Foundry; Agent, Plan, Ask, Edit modes in VS Code, JetBrains, Eclipse, Xcode |
| January 2026 | Support for any OpenAI-compatible provider + AWS Bedrock and Google AI Studio; models with Responses API |
| June 23, 2026 | BYOK in the GitHub Copilot app — public preview for all paid plans |
| July 14, 2026 | Custom OpenAI-compatible endpoints in Copilot for JetBrains (Pro plans and above) |
Connecting JoinGonka Gateway: step-by-step
You will need a jg-… key — it is issued upon registration on the gateway along with 10M free starter tokens; no credit card is required.
| 1. Settings | Open Settings → Model Providers (in VS Code — via Copilot Chat, in the Copilot application — in account settings) |
| 2. Provider | Add a provider of type "OpenAI compatible" with the endpoint https://gate.joingonka.ai/v1 |
| 3. Key | Paste your jg-… — it will be saved in your local keychain |
| 4. Model | Network models will appear in the switcher next to the built-in ones; the current list is GET /v1/models: moonshotai/Kimi-K2.6, MiniMaxAI/MiniMax-M2.7, deepseek-ai/DeepSeek-V4-Flash-0731 |
| 5. Session | Choose a model for each session — Copilot-hosted and BYOK models can be switched freely |
Tip. The command npx @joingonka/setup --tool copilot-byok will print ready-to-use values for the fields below (base URL, key, model ID with current limits) and perform a live request to verify that the gateway accepts them. Copilot itself is configured only via the UI — it has no file that can be safely edited, so you must insert the values manually.
Which network model to choose
All three network models support streaming and tool calling. deepseek-ai/DeepSeek-V4-Flash-0731 has the network's longest context (380K tokens) and an output of up to 32768 tokens: use it for large repositories and long agentic sessions. moonshotai/Kimi-K2.6 is a strong agentic model for multi-step tasks. MiniMaxAI/MiniMax-M2.7 is universal; Kimi and MiniMax support outputs of up to 8192 tokens.
Public preview limitations — honestly
BYOK covers agentic sessions and chat; inline suggestions in the editor may continue to run on Copilot models — this is a limitation of Copilot itself, not the provider. The Enterprise BYOK variant is configured at the organization level and is available for GitHub Enterprise/Business. Vision tasks do not work through the gateway: the network models are text-based. If a tool sends requests in the Responses API format, it is not currently supported on the gateway; please select Chat Completions.