Collection

Models and infrastructure

Model selection, inference economics, local deployment, compression and serving architecture.

Blog posts
80
Topic
AI and agents

Start here

Latest in this collection

Two overlapping waveforms illustrate conversational timing in voice AI AI & Agents

Cue Turn Detection: Phone Numbers, Benchmarks and Limits

A pause is not a finished phone number. What Pype’s Cue v1 improves, what the benchmarks measure, and how to test correct, confirmed capture.

Kevin Riedl
Read
Wavect editorial header for Agent Mesh, with a shield representing controlled access to tools AI & Agents

Google SAM is now Agent Mesh: MCP tools across machines

Connect approved tools across machines. Understand the control plane, signed EU labels and the checks a useful first pilot should pass.

Kevin Riedl
Read
Layered document cards representing EdgeQuake's PostgreSQL knowledge store AI & Agents

EdgeQuake: PostgreSQL GraphRAG for MCP agent memory

Connect document relationships to your agents. A v0.34.0 review of PostgreSQL deployment, MCP access, local model roles and the LightRAG benchmark.

Kevin Riedl
Read
Three ternary weight values, minus one, zero and plus one, connected to a CPU AI & Agents

BitNet on CPU: 2B4T, real RAM needs and the 100B claim

Which BitNet model can you download, and what fits your CPU? Separate the 100B benchmark, 0.4 GB metric and 1.19 GB file, then test a real task.

Kevin Riedl
Read
SuperPenguin editorial illustration about AI cost attribution per pull request, customer and feature AI & Agents

SuperPenguin: AI Cost per PR, Customer and Feature

What did a merged PR cost, and what does each customer cost to serve? A practical SuperPenguin guide with billing reconciliation, worked examples and an optimization prompt.

Kevin Riedl
Read
Editorial illustration of product dashboard findings becoming an animated update AI & Agents

Claude Dashboards + Motion: From KPIs to Video

Turn a reviewed product metric into a short animated update. A practical Claude Dashboards and Motion workflow with a cohort example, two prompts and sharing rules.

Kevin Riedl
Read

Complete article directory

  1. Cue Turn Detection: Phone Numbers, Benchmarks and Limits
  2. Google SAM is now Agent Mesh: MCP tools across machines
  3. EdgeQuake: PostgreSQL GraphRAG for MCP agent memory
  4. BitNet on CPU: 2B4T, real RAM needs and the 100B claim
  5. SuperPenguin: AI Cost per PR, Customer and Feature
  6. Claude Dashboards + Motion: From KPIs to Video
  7. OpenAI Intelligent UI vs OpenUI: Generative UI for Business Apps
  8. Wally by RunAnywhere: Cost per Accepted Coding Task
  9. Reducto for German Invoices: Measure Human Review Effort
  10. OpenAI Decisions API: Confidence, Refusals and Routing
  11. Claude Model Router: What Switches the Model, and When?
  12. Cloudflare Clef vs Jev: Pricing, Benchmarks and Migration
  13. Context Language Models vs Compaction: What to Pilot
  14. Caveman 3.0 for Claude Code: Local Input Compression, Recovery and Benchmarks
  15. LiteLLM Lens: Agent Trace Analysis with SQL and APIs
  16. SmythOS Studio Self-Hosting: Docker Setup, Costs and Limits
  17. DeerFlow 2.0: Docker Setup, Sandboxes and Memory
  18. CLM-8B Self-Hosting: vLLM, Action Cache and Verifiers
  19. AnyJev: LLM Calibration and Option-Order Bias
  20. LiteAgents SDK: Per-Turn Routing, Setup and Migration
  21. mcp-memory-service: Shared Memory for Claude Code and Cursor
  22. Claude Opus 5.5: Best Uses, Prompts and Effort Settings
  23. Laya and Jev for Business: 6 Practical Workflows
  24. Laya vs Jev: What the Benchmarks Mean for AI Startups
  25. Jev AI Review: Decision Models for Agent Workflows
  26. SwarmLLM Review 2026: Browser P2P LLM Inference Across Phones and Laptops
  27. Phonely Alma Review: Is the Voice LLM Ready for Production?
  28. Utopia Review: Temporal Knowledge Graph for Enterprise
  29. NVIDIA PAIR Review: Local AI Routing and the AMD Gap
  30. Tencent Hy4 Preview Review: Is the 1M-Context Coding Model Worth a Pilot?
  31. PageLM Review: Self-Hosting and Commercial Use
  32. M6 Mac mini vs M5 Mac Studio for Local AI: Buyer's Guide
  33. FreeToken AI Review: Run Frontier MoE Models on Consumer GPUs?
  34. Darkbloom AI Review: Private Inference on Idle Macs
  35. Ox Alpha Revealed as GLM-5.3-Flash: What Changed
  36. Pika Audio Models API Pricing: Is SFX Really 20x Cheaper?
  37. Thunder Compute's $13M GPU Virtualization Bet: Enterprise Buyer's Guide
  38. AirLLM on 4 GB VRAM: How Layer-Wise Inference Really Works
  39. Qwen3.8-27B: Self-Hosted Computer-Use Agents Without Exporting Screenshots
  40. Netflix's vLLM and Triton Stack: 7 Production Lessons
  41. Transformers.js Browser AI: When Local Inference Belongs in Your Product
  42. How to Self-Host LiteLLM in Production: 2026 Guide
  43. AI-Ready Company Wiki: Architecture and Build Guide
  44. Does Claude Watermark Text? The 2026 API Answer
  45. OpenKB Review: Knowledge Compiler vs RAG
  46. Unsloth Desktop Review: A Private Local AI Workstation?
  47. NeMo Switchyard 0.2: Agent Model Routing Without Training?
  48. Firecrawl AnyDoc Review: 14 Formats to Markdown
  49. Muse Glimmer 30B: Is Meta's Local Agent Model Production-Ready?
  50. OmniRoute AI Routing: Setup and Production Checklist
  51. Gemini Robotics 2: Whole-Body Control and the Pilot Decision
  52. pdf-inspector Review: Route PDFs Before OCR
  53. Local Multimodal AI Coding Assistant: Voice, OCR and Privacy
  54. DeepSeek V4 Flash 0731 on One AI PC: What Actually Works?
  55. Fine-Tune Gemma 4 Free with Unsloth and Colab
  56. Taalas HC1 Review: Is a Hardwired LLM ASIC Worth It?
  57. llmfit Guide: Which Local LLM Fits Your Hardware?
  58. Miso TTS Self-Hosted vs API: Cost, Latency and VRAM Reality
  59. Compress RAG Vectors to 2 Bits: Is Data-Oblivious Quantization Production-Ready?
  60. LMCache Reported 13.7x Lower Mean TTFT with a Shared KV Cache
  61. Lossless LLM Weight Compression vs 8-bit GGUF: What Is Ready for Production?
  62. NVIDIA Nemotron 3.5 ASR: Is Free Self-Hosted STT Ready for Voice Agents?
  63. Kimi K3 for EU Companies: API Cost, Data Risk, and a Pilot Plan
  64. Mesh LLM Review: Can One Large LLM Run Across Multiple Computers?
  65. Bonsai 27B Review: Can a 27B LLM Really Run on a Phone?
  66. Soofi S: Is Germany's Sovereign LLM Ready for Business?
  67. Colibri Runs GLM-5.2 on Consumer Hardware. Here Is the Catch.
  68. When Local Models Beat APIs: A Break-Even Calculator for EU Companies
  69. LLM Cost Calculator 2026: Cost per Task, Not Cost per Token
  70. Pxpipe Review: Can Images Cut Claude Code Token Costs 60%?
  71. Cheaper Per Token Can Still Cost More Per Task
  72. Anthropic Says No Open-Weights Ban. The Cost War Is Still Real.
  73. What an Internal AI Assistant Actually Costs in the DACH Region (2026)
  74. LLM Gateways Compared 2026: LiteLLM vs OpenRouter vs Portkey vs RouteLLM
  75. Self-Hosting LLMs in the EU: When Open Weights Actually Pay Off
  76. Open-Weight LLMs 2026: DeepSeek, Qwen, Kimi, GLM, Llama and Muse
  77. How to Cut LLM Token Costs in 2026
  78. When Is an LLM Eval Worth Building? Cost, ROI, and Trusting the Judge
  79. RAG vs Fine-Tuning vs Long-Context 2026
  80. LLM API Costs 2026. Architecture Shift
Inbox, without the noise

Get the next AI and agents field note

One concise email when we publish. No tracking pixels, and no inbox filler.