Blog
Use SolCard Inference in Cursor with a custom base URL
Point Cursor’s Override OpenAI Base URL at SolCard Inference: the settings, the model IDs, Agent vs Ask mode, and what still passes through Cursor’s servers.
6 min readUse SolCard Inference with the OpenAI Agents SDK
Run OpenAI Agents SDK agents on SolCard Inference: the custom client, why to use Chat Completions, turning tracing off, and which SDK features to skip.
6 min readUse SolCard Inference in the Pi coding agent
Add SolCard Inference to Pi with one models.json provider: the api type, compat flags, context and token limits, a check, and what a turn costs.
6 min readUse SolCard Inference in Zed
Add SolCard Inference to Zed’s Agent Panel as an OpenAI-compatible provider: the settings.json block, the key variable, reasoning_effort values and fixes.
6 min readUse SolCard Inference with LangChain
Point LangChain’s ChatOpenAI or ChatAnthropic at SolCard Inference: the base_url for each, streaming usage, reasoning, and the two features to leave off.
5 min readUse SolCard Inference in Crush
Add SolCard Inference to Charm’s Crush with the new crushrc file: the provider add and model add lines, context and price flags, a check, and fixes.
6 min readUse SolCard Inference with the Vercel AI SDK
Call SolCard Inference from TypeScript with the AI SDK’s OpenAI-compatible provider: the createOpenAICompatible setup, streaming usage, reasoning and costs.
5 min readUse SolCard Inference in Goose
Add SolCard Inference to Goose as a custom OpenAI-compatible provider: the Desktop form, the custom_providers JSON file, token limits, and what a session costs.
5 min readUse SolCard Inference in Continue
Add SolCard Inference to Continue in VS Code or JetBrains: the config.yaml block, where the key goes, which model to pick, and what a chat turn costs.
5 min readUse SolCard Inference in Kilo Code
Add SolCard Inference to Kilo Code as a custom OpenAI Compatible provider: the dialog fields, the kilo.jsonc block with token limits, and what a task costs.
5 min readUse SolCard Inference in VS Code with Custom Endpoint
Add SolCard Inference to VS Code chat with the built-in Custom Endpoint provider: the chatLanguageModels.json block, which model to pick, and what a turn costs.
6 min readAbliterated, uncensored, jailbroken: three different things
Abliteration edits a model’s weights, uncensored fine-tunes retrain it, a jailbreak is only a prompt. How each works, what it can’t do, and hosted vs local.
5 min readRun Aider on SolCard Inference
Point Aider at SolCard Inference with OPENAI_API_BASE and —model openai/: the exact commands, the unknown-model warning, a check, and what an edit costs.
3 min readPoint Claude Code at SolCard Inference with two environment variables
Set ANTHROPIC_BASE_URL and ANTHROPIC_AUTH_TOKEN and Claude Code runs on SolCard Inference. The exact values, how to check them, and what you give up.
4 min readCan you paste client code into an AI tool? A checklist for freelancers
Pasting client code into ChatGPT or any AI tool raises two questions: what your NDA allows, and what the provider keeps. A checklist for both.
5 min readSet up SolCard Inference in Cline
Add SolCard Inference to Cline as an OpenAI Compatible provider: base URL, key, model id, the Model Configuration values, and what a task costs.
4 min readSolCard Inference vs Featherless: flat rate or pay per token
Featherless runs 40,000+ open models on a flat chat plan or per-token credits. SolCard Inference runs three uncensored ones. Price, privacy, payment compared.
6 min readUse SolCard Inference in Open WebUI and LibreChat
Connect Open WebUI or LibreChat to SolCard Inference’s OpenAI-compatible API: the URL, the key, the models, and where your chat history ends up.
4 min readUse SolCard Inference in OpenCode
Add SolCard Inference to OpenCode as a custom OpenAI-compatible provider: the opencode.json block, which model to pick, a check, and what a session costs.
4 min readPay for an AI API with USDC: no card, no expiry
Fund SolCard Inference’s API with USDC or other crypto: no card on file, a balance that never expires, and what to do if you pick the wrong network.
4 min readFast, Large or Base: which SolCard Inference model to use
Three models, one price list. What each is good at, what a request costs on each, and when turning on reasoning is worth it.
3 min readAdd SolCard Inference to Claude Code and Codex in one sentence
Paste one line into your coding agent and our models show up next to your own. What the setup changes, how to switch, and what a session costs.
3 min readWhere your balance goes: how SolCard Inference bills a request
One prepaid balance, a hold before each request and a charge after it. What a request costs, what a plan covers, and what happens when the money runs out.
4 min readSolCard Inference vs OpenRouter: which one fits you
OpenRouter routes you to hundreds of models from many providers. SolCard Inference runs three uncensored open models with nothing kept. How they differ on privacy, price, payment and coding agents.
4 min readWhat makes an LLM API private? Seven questions to ask
“We don’t train on your data” isn’t the same as private. Seven questions that show what an AI provider really keeps, and how SolCard Inference answers each one.
3 min readSolCard Inference vs Venice: two private, uncensored AI services compared
Both keep your chats on your device and offer uncensored models. Where they differ: which requests are private, what a plan covers, how you pay, and how they plug into coding agents.
4 min readWhat is SolCard Inference?
A chat app and an API for open models that answer what other assistants refuse, with nothing you write kept anywhere. Prepaid, paid in crypto, from the team behind SolCard.
3 min readKeeps nothing: follow one message through SolCard Inference
One ordinary message, traced from your keyboard to the model and back: everything that touches it, for how long, and what’s left when the answer lands. Nothing you wrote.
7 min read