Category
AI Development Articles
Deep dives into AI-assisted and agentic development. Coding agents, frontier model releases, SDKs, prompting patterns, and the engineering workflows behind building production software with AI.
For current model prices, access limits, and workload selection, use our maintained frontier model guide.
Latest Articles
The newest AI Development guides and analysis
OpenRouter's full catalogue compared across October 28 to November 5, 2026: 665 matched routes, 35 price changes, 33 additions, 27 delistings, 24 expiry dates.
#ai-api-pricing#openrouter+4 more
2026-11-05
Read Article
We sent an unauthenticated initialize to 1,000 remote MCP servers sampled from the official registry on Nov 5, 2026: 55% answered, 34% asked for credentials.
#mcp#model-context-protocol+4 more
2026-11-05
Read Article
OpenAI shipped GPT Image 2.6 on November 4, 2026 with GIF and SVG output, markup edits and a complexity dial. How Flare and Sunburst differ and what they cost.
#gpt-image-2.6#openai-images+4 more
2026-11-04
Read Article
Seven image models compared on November 5, 2026: prices in each vendor's own units, two independent leaderboards, edit features and what is not comparable.
#image-generation-models#gpt-image-2.6+4 more
2026-11-04
Read Article
Ricursive released Ripple-65B Preview on November 4, 2026: a 65B dense model, 1M context, Apache 2.0. What the vendor claims, what is verifiable, what it costs.
#ripple-65b#ricursive+4 more
2026-11-04
Read Article
Tencent released Hy4 on November 4, 2026 with vision and computer use, at the preview's API price. The weights are promised, not published. What is confirmed.
#tencent-hy4#hunyuan+4 more
2026-11-04
Read Article
Gemini 4 Carbon, GPT-6 Rosalind Research and Claude Mythos 5.1 ship only to vetted organisations. Who qualifies, how to apply and what each vendor states.
#Gemini 4 Carbon#GPT-6 Rosalind+4 more
2026-11-03
Read Article
Anthropic's November 2 S-1/A states $33.2B nine-month revenue, $895.4B of compute commitments and capacity risks. What the filing tells Claude API buyers.
#Anthropic#S-1+4 more
2026-11-02
Read Article
Anthropic's Messages API can now hold a conversation server-side until 5 minutes after the last request. What a thread stores, costs and where it is barred.
#Claude API#message threads+4 more
2026-11-02
Read Article
The prompting mistakes each vendor's own guide names, quoted: thinking prompts on reasoning models, shouting, prefills, examples, and where the three disagree.
#prompt-engineering#openai+5 more
2026-11-01
Read Article
Seven failure points from Barnett et al., the lost-in-the-middle effect, a 16-type error taxonomy and Anthropic's vendor-run retrieval figures, limits stated.
#rag#retrieval-augmented-generation+5 more
2026-11-01
Read Article
OpenAI shuts down gpt-5, gpt-5-mini, gpt-5-nano, gpt-5-pro, o3 and o3-pro on December 11, 2026. The snapshots, the named replacements and what the switch costs.
#openai#gpt-5+5 more
2026-10-31
Read Article
Eleven MCP server design mistakes drawn from the 2026-07-28 spec, its security best practices and Anthropic's tool guidance, with the rule each one breaks.
#mcp#model-context-protocol+5 more
2026-10-31
Read Article
Anthropic deprecated fast mode on Claude Opus 4.8 on October 30; from November 30 those requests error. The exact wording, the replacement and a checklist.
#Claude API#Anthropic+3 more
2026-10-30
Read Article
Anthropic's Usage and Cost Analytics API launched in beta on October 29. What its two endpoints return, who can call them and how to leave the legacy APIs.
#Claude API#Anthropic+3 more
2026-10-30
Read Article
NVIDIA's Nemotron 3.5 Transcribe, released October 28, against Whisper large-v3 and Parakeet TDT 0.6B v3 on licence, languages, size, hardware and WER.
#Speech to Text#NVIDIA+3 more
2026-10-30
Read Article
What BFCL, τ²-bench, MCP-Universe, MCPMark and Toolathlon each measure, how they score, who leads them as of November 5, 2026 and how to read a launch claim.
#AI Benchmarks#Tool Calling+3 more
2026-10-29
Read Article
GitHub shipped Copilot Fusion model routing and a Max code-review level on October 28. What each routes, which plans get it, admin policies and credit spend.
#GitHub Copilot#AI Coding+3 more
2026-10-29
Read Article
Mistral released Large 4 weights on October 29 and lists $0.70 input and $2.10 output per million tokens. Direct prices compared with DeepSeek, Kimi and GLM.
#AI Models#Mistral+3 more
2026-10-29
Read Article
Google announced Gemini 4 Carbon on October 28, with access phased to cyber defenders via Fairwind. Who qualifies, the announced price and what to do meanwhile.
#AI Models#Gemini+3 more
2026-10-29
Read Article
Tencent’s Hy-SimulMT reports roughly three seconds of translation lag. Understand the metric, Chinese-English scope, TTS dependency and deployment limits.
#Hy-SimulMT#Tencent Hunyuan+4 more
2026-10-28
Read Article
LFM3 brings text, vision, audio and decision models to local devices. Compare the variants, memory budgets, licence limits and tasks worth testing first.
#LFM3#Liquid AI+4 more
2026-10-28
Read Article
We compared 680 OpenRouter routes across 26–28 October: 24 pricing records changed. Explore every change, alias caveats and downloadable reproducible data.
#AI API pricing#OpenRouter+4 more
2026-10-28
Read Article
Unlimited-OCR 2 targets very long documents with bounded attention state. Test cross-page accuracy, memory, provenance and review costs before deployment.
#Unlimited-OCR 2#document parsing+4 more
2026-10-28
Read Article
Digital Applied newsletter
Deep dives on AI, marketing and development.
Practical guides and fresh insights by email. No recycled takes.