LLM API

12 articles in this category.

GPT Latest Model: What ~openai/gpt-latest Points At Now
LLM API

GPT Latest Model: What ~openai/gpt-latest Points At Now

As of September 8, 2026, the latest GPT model is GPT-6 Astra (gpt-6-astra): what ~openai/gpt-latest points at now, what the switch cost, and when to pin.

MiniMax M3 API: Integration Guide, Pricing & Code
LLM API

MiniMax M3 API: Integration Guide, Pricing & Code

MiniMax M3 API guide: model ID, 1M-token context, official per-million pricing as of September 2026, both base URLs, working curl and SDK code, and M3 vs M2.7.

Agent Observability: What Your Coding Agent Called and Spent
LLM API

Agent Observability: What Your Coding Agent Called and Spent

Choose the right agent observability layer: gateway request logs show every model call, token count, and cost per end-user, but not tool calls or agent loops.

Anthropic Base URL: Claude Code and Harness, No /v1
LLM API

Anthropic Base URL: Claude Code and Harness, No /v1

Set ANTHROPIC_BASE_URL to https://llm.modellix.ai, never /v1: Claude Code settings.json and the harness anthropic-messages provider, plus the 401 auth traps.

OpenAI Compatible API: What 'openai-completions' Means
LLM API

OpenAI Compatible API: What 'openai-completions' Means

What 'openai-completions' means: an OpenAI compatible API is a protocol contract, not a switch. Changing only the URL breaks; the cheat sheet shows why.

DeepSeek Harness API Key: Which Key Goes Where
LLM API

DeepSeek Harness API Key: Which Key Goes Where

One Modellix API key powers LLM, Design, and Web in DeepSeek Harness. Where it lives depends on how you add it: $DSH_HOME/.credentials.yaml or MODELLIX_API_KEY.

DeepSeek Harness Provider Errors: UNKNOWN_MODEL and 401s
LLM API

DeepSeek Harness Provider Errors: UNKNOWN_MODEL and 401s

Decode DeepSeek Harness provider errors: UNKNOWN_MODEL means the models list is the catalog, MISSING_CREDENTIAL means no key, only a 401 marks a key invalid.

DeepSeek Harness Token Usage: What an Agent Run Costs
LLM API

DeepSeek Harness Token Usage: What an Agent Run Costs

See what one DeepSeek Harness agent run costs: query GET /v1/logs for per-request cost, token buckets, and session attribution, then convert per 1M tokens.

DeepSeek Harness Plugin: Bundles and When You Don't Need One
LLM API

DeepSeek Harness Plugin: Bundles and When You Don't Need One

DeepSeek Harness plugins ship as bundles: a package manifest, a patch file, a profile entry. Here is how bundles work — and when pure configuration is enough.

DeepSeek Harness Image Generation: Media Output via Plugin
LLM API

DeepSeek Harness Image Generation: Media Output via Plugin

Generating images inside DeepSeek Harness requires a plugin: the Design module adds image, video and audio output, and the billing boundaries most guides skip.

dsh plugin add: Install, Verify, Remove Harness Bundles
LLM API

dsh plugin add: Install, Verify, Remove Harness Bundles

dsh plugin add installs a bundle into a profile, but installed does not mean effective. Verify with dsh --dump-config, pack a tarball, or remove cleanly.

DeepSeek Harness Web Search: web_search and web_fetch
LLM API

DeepSeek Harness Web Search: web_search and web_fetch

Swap DeepSeek Harness web_search and web_fetch providers: install dsh-modellix, add a Modellix key, and both native tools run through tool.modellix.ai.