The Ollama API and port 11434 explained
Curl your Ollama server on port 11434, read what a refused connection means, tour the API endpoints, and set OLLAMA_HOST without exposing the box.
Filtering by topic #api · clear
Curl your Ollama server on port 11434, read what a refused connection means, tour the API endpoints, and set OLLAMA_HOST without exposing the box.
Four ways to authenticate a Claude API client on a VPS: an Anthropic key, AWS IAM on Bedrock, Google ADC on Vertex, Entra on Foundry, plus safe storage.
Run Codex, Claude Code and Hermes behind one self-hosted API. The exact Docker deploy, the loopback bind, the default login you must change, and TLS access.
Mock servers and API test runners do different jobs. Run both on your own box: WireMock stubs in git, Hurl suites in CI, and reports that survive a rebuild.
The Ollama server ships with no authentication, so anything that reaches port 11434 can run your models and pull new ones. The three fixes, in order.
Claude output tokens cost five times input. Here is why decoding is slower than prefill, and what that asymmetry does to a real agent bill each month.
Output tokens cost five times input, and every turn re-reads your whole context. See the arithmetic on one real session, and the four levers that cut it.
Cache writes cost 1.25x and reads 0.1x, so a Claude prefix pays back on its second use. Derive your own break-even, then prove it from the API.
No free tier, a small credit at signup, and several parts that cost nothing at all. The July 2026 rates, plus a worked cost for a first small app.
Switching models does not restore access. How Claude subscription session and weekly limits differ from API 429 rate limits, and what to do next.
An unattended agent bills every loop with nobody watching. Hard caps, task budgets, prompt caching, batching, and the usage fields that show the spend.
Use Claude to analyze stocks and options with live market data: a free no-key API, three ways to connect it, and copy-paste prompts that get grounded answers.