Coding agents
How to Use Codex in Cursor
Install OpenAI’s Codex extension in Cursor, sign in with an eligible ChatGPT subscription or OpenAI API key, and distinguish the extension from Cursor’s built-in model settings.
AI model and infrastructure insights
Insights, page 44 of 95
Model APIs
GLM-4.7-Flash on Ollama: How to Run It and When to Use an API
Run GLM-4.7-Flash with Ollama, evaluate its memory use and response time at your intended context length, and compare local use with a production API.
Image generation
How Many Images Can You Generate With Gemini Free?
Gemini doesn’t publish a fixed number of free images. Learn how its usage limits work, where to find your refresh notice, and how API access differs.
Model APIs
DeepSeek Janus-Pro: Capabilities, API Access, and Model Fit
Learn what Janus-Pro is suited for, what DeepSeek’s current API documentation lists, and whether a Janus-Pro route is publicly documented by Token Forge Cloud.
Model APIs
Gemini 2.5 Flash API key: get a key and make your first API call
Create a Gemini API key, check whether your account can use Gemini 2.5 Flash, and send a direct REST request with the correct model ID.
Image generation
How Many Images Do You Get With ChatGPT Plus?
As of November 9, 2026, the OpenAI Plus help material checked for this article does not give a fixed ChatGPT image-generation count or reset interval. See how that differs from image uploads and when an API may fit your workload.
Pricing and plans
Grok 4 API Pricing: What Legacy grok-4-0709 Calls Cost
The original Grok 4 model is retired from xAI’s API. Learn what happens to legacy grok-4-0709 requests, the replacement rates, and how to estimate API spend.
Pricing and plans
Is Gemini 2.5 Pro Free? Access and API Pricing Limits
Gemini 2.5 Pro is not confirmed as free across the app, AI Studio, CLI, or API. See what current documentation says about access, the December 1, 2026 API notice, and what remains unknown about exact API costs.
How-to guides
How Many Parameters Does GPT-4 Have?
OpenAI’s GPT-4 technical report does not disclose a parameter count. Learn why the numbers online are estimates—and why they don’t determine API suitability.
Model APIs
Gemini Deep Think API: Argon Integration and Availability
Learn what Google documents about Gemini Deep Think and API access, then follow a Gemini 4 Argon REST example with API-key and billing setup.
Pricing and plans
How Often Do Claude Limits Reset? Free, Paid, and API
Claude Free usage resets every five hours; most paid plans also have session limits and weekly limits. See where to check your reset time and how API rate limits differ.
Video generation
How to Get Sora 2 on Android: Current Status and Safe Next Steps
Sora is no longer available: OpenAI discontinued its web and app experiences on April 26, 2026, and set September 24, 2026 as the Sora API shutdown date. Here’s what Android users should know.
Pricing and plans
Grok 4 Pricing: API Rates, Legacy Model IDs, and Cost Examples
The original Grok 4 API model was retired and its legacy ID redirects to Grok 4.3. Compare current API rates, estimate cached-token and tool costs, and distinguish API billing from subscriptions.
Video generation
Google Veo 3.1 Pricing: Current API Access and Historical Rates
Google’s Gemini API Veo 3.1 preview models were shut down on October 22, 2026. See what Google documents about migration and production access, and why the old API rates are not a current quote.
Model comparisons
ChatGPT Desktop App vs. Browser: Which Should You Use?
Compare ChatGPT on desktop and in a browser, including local-computer workflows, eligibility caveats, and when a production app needs the OpenAI API.
Pricing and plans
Is GLM-4.6 Free? API Pricing and Production Access
GLM-4.6 is not listed as a free model on Z.ai’s API price sheet. See its per-token rates, a sample cost calculation, and the key distinction between free cache storage and free model usage.
Pricing and plans
Is Claude Free Better Than ChatGPT? Compare Plans and API Costs
Compare Claude Free and ChatGPT Free by supported features and usage limits, then see how production API billing and token costs differ from consumer plans.
Image generation
How Many ChatGPT Images Can You Generate Per Day?
OpenAI’s reviewed help pages don’t establish a fixed daily image count for ChatGPT. See what they say about usage limits and how to assess image-generation capacity through the API.
Image generation
How Long Does ChatGPT Take to Create an Image?
OpenAI says image generation may take a few minutes, depending on request complexity. Learn what that means, how to troubleshoot a stalled request, and why ChatGPT wait times don’t predict API latency.
Coding agents
How to Build Skills in Claude: A Step-by-Step Tutorial
Build a reusable Claude Skill by writing a focused SKILL.md, adding it to Claude or Claude Code, and testing it against real tasks. Includes a Skill-specific API example.
Coding agents
Cursor API Pricing: Cloud Agents, Subscription Plans, and Usage Costs
Cursor’s Cloud Agents API runs coding-agent workflows, not general model inference. See subscription pricing, how included usage works, and what to confirm before budgeting an API workload.
Model APIs
Gemini 3 for Coding: API Models, Benchmarks, Pricing, and How to Evaluate
Learn how to select a Gemini 3 coding model, run a documented API smoke test, verify production access and data-handling requirements, and evaluate it on your own code.
Audio and speech
Whisper AI Model: What It Does and How to Choose an API
Learn what OpenAI Whisper can do, when to choose offline transcription versus live streaming, and what to verify before selecting local inference or a production API.
Video generation
How Much Is Google Veo? Flow Credits and Plan Prices
Google Veo has free and paid Flow access, while production API costs depend on the route. Compare U.S. plan prices, Flow credits, and current API limitations.