Research library
Software engineering
Practical architecture and development decisions for software that people can use, maintain, and adapt.
108 articles · Page 1 of 9
Claude's Usage and Cost API Goes Legacy: What Cost Dashboards Gain and Lose
Claude's Usage and Cost Analytics API adds actor and cost-type grouping, but data typically lags up to 4 hours and ranges cap at 31 days. Plus 128 MB requests.
How the New Flash Models Work: Inside GLM-5.3-Flash and Qwen3.8-Flash-Next
GLM-5.3-Flash and Qwen3.8-Flash-Next cut serving cost with hybrid linear and sparse attention, widened residual streams, MTP and host-memory n-gram tables.
Gemini Computer Use Is GA: A Builder's Checklist Against GPT-6.1 Terra
Gemini computer_use is GA. Which model to pick, three environments, the intent field, two safety traps, and what it costs next to GPT-6.1 Terra.
Gemini API Now Stores Requests From New Paid Projects for 55 Days by Default
Since October 26, new paid-tier Gemini projects log generateContent requests for 55 days unless store=false. What the toggle covers, and what it does not.
GitHub Agentic Workflows Go GA, and Copilot Apps Can Bill the Org Without a Seat
Agentic Workflows is GA: Markdown to Actions YAML, read-only agents behind a firewall, 1,000 AIC per run. Server-to-server auth bills apps to the org.
Xinference 4.1.0: CosyVoice RCE Patched, First-Admin Takeover Closed, Actor Auth Still Off
Xinference 4.1.0 fixes CosyVoice YAML code execution and an unauthenticated first-admin claim. Upgrading sets neither the trust flag nor XOSCAR_AUTH_KEY.
Anthropic's Five Exploit Chains: The Patch Checklist First, Then the AI Story
Fixed versions for Chrome 156, Linux CVE-2026-98432 and Windows CVE-2026-33096 from Anthropic's exploit-chain post, and where the CVE records disagree.
ChatGPT Pro and Business Plans Now Carry Up to $500 a Month in API Credits
ChatGPT Pro plans now get $100 to $500 a month in API credits, Business $20 or $100 a seat. Eligibility, exclusions, the 429 at zero and what $200 buys.
Copilot Fusion Bills Per Model, and GitHub Enterprise Meters Copilot From the First Request
Copilot Fusion is GA: single, cascade or critique routing billed per model, 30% off to Nov 30. GitHub Enterprise now meters Copilot with no credit pool.
Gemini 4 Carbon Goes to Cyber Defenders First, at Argon's Price
Gemini 4 Carbon starts with cyber defenders and Fairwind partners, not the API. Its intro price matches Argon's $2/$10 with no stated end date.
Liquid LFM3: Six 2B-Class Open-Weight Models, One Revenue-Capped License
Liquid's LFM3 ships text, vision, audio, omni and two decision models near 2B parameters under the LFM Open License. Sizes, memory, runtimes, $10M line.
GPT-6.1 Terra Halves Sol's Price for High-Volume Computer Use
GPT-6.1 Terra costs $1/$5 per million tokens, half of GPT-6.1 Sol. Specs, migration limits, vendor benchmarks, Preparedness results and what to test.