OpenAI's GPT-6.1 Terra replaces GPT-5.6 Terra at $1/$5 per million tokens. The specs, the price cut that's bigger than the headline, every benchmark with its source, and when to pick it over Sol or Luna.

Ashley Innocent

October 27, 2026

Kimi K3.1 Pro is Moonshot AI's post-trained Kimi K3: same 2.8T MoE design, 1M-token context, one third cheaper at $2/$10 per 1M tokens, and far fewer hallucinations on Moonshot's tests. Specs, vendor-run benchmarks, pricing, Highspeed and the open-weights date.

Ashley Innocent

October 24, 2026

Two critical PHP bugs, CVE-2026-109618 and CVE-2026-109619, give unauthenticated attackers code execution on every SAPI that serves HTTP. Here's how to check your servers, patch to 8.5.12/8.4.27/8.3.36/8.2.35, and what the workarounds really cover.

Ashley Innocent

October 23, 2026

Google released Gemini 4 Argon without cyber guardrails to Fairwind defenders, and since October 13 any team with a paid Gemini API key can call gemini-4-argon. Here are the cyber numbers, both access routes, Google's safeguards, and the API checks to schedule now.

Ashley Goolam

October 15, 2026

OpenAI's Decisions API (POST /v1/decisions, GPT-6 Luna, GA since Oct 22, 2026) turns text, images and JSON into predicate, choice and score answers with probabilities, at $0.10 per 1M input tokens ($0.01 cached) and no output charge. What it returns, costs, and when to pick it.

INEZA Felin-Michel

October 10, 2026

Decisions for classify, route, score and gate with probabilities and input-only billing at $0.10 per 1M; Responses for text, your own JSON, tools, streaming and state. The same ticket routed both ways, the cost math, a feature matrix and testing both in Apidog.

INEZA Felin-Michel

October 10, 2026

OpenAI's Decisions API (GA since Oct 22) and TypeSafe's Jev both return typed answers with probabilities and bill input only. Compare price, caching, inputs, limits, access and output shapes, with the same ticket-routing request in both, then test both in one Apidog project.

Ashley Innocent

October 10, 2026

What is Claude Haiku 5.5? Anthropic's Oct 7, 2026 model: $0.10/$0.50 for prompts up to 100K tokens, 1M context, benchmarks, and changes from Haiku 4.5.

Ashley Innocent

October 8, 2026

Nano Banana 2.1 (gemini-nano-banana-2.1) costs half of Nano Banana 2 per image. What Google says improved, full pricing, and the catches.

Ashley Innocent

October 6, 2026

Mistral Large 4 scores 82% on a cyber test where GPT-6 Astra and Claude Opus 5.5 score near zero. Here's why, plus specs, $0.68 pricing and where it loses.

Ashley Innocent

October 6, 2026

Gemini 4 Argon is GA on the paid Gemini API and leads Google's table on 13 of 19 rows. Here's how it compares with GPT-6 Astra and Claude Opus 5.5 on price, limits and benchmarks, which to route each task to, and how to test all three.

Ashley Innocent

October 2, 2026

Gemini 4 Argon costs $2/$10 per 1M tokens through November 30, 2026, then $4/$20, with cached input 95% off. Here's the full rate card, cost math for real workloads, a maxed-out long decode, and how it compares with Opus 5.5, Astra and Sol.

INEZA Felin-Michel

October 2, 2026

Gemini 4 Argon is Google's frontier model, generally available since October 13, 2026 as gemini-4-argon on the paid Gemini API and for Pro and Ultra subscribers in the Gemini app. Here's what it costs, how it benchmarks, and how to call it.

Ashley Innocent

October 2, 2026

MCP Events explained: how ChatGPT subscribes to your MCP server with events/subscribe, Standard Webhooks signing, the 256 KiB limit, and tests in Apidog.

GPT-6.1 Sol API guide: your first gpt-6.1-sol request, effort levels, Batch/Flex/Fast pricing, and the four changes to migrate from gpt-6-sol.

INEZA Felin-Michel

September 30, 2026