Claude Sonnet 5.5 Explained: Pricing, Benchmarks, Terminal-Bench
Anthropic shipped Claude Sonnet 5.5 on Sept 28, 2026 at $2/$10 per million tokens. What changed vs Sonnet 5, sourced benchmarks, and when to pick it.
Xiaomi MiMo-V2.6 Explained: MIT-Licensed Open Weights and an Open RL Stack
Xiaomi released MiMo-V2.6 Pro and Flash under MIT license with its RL training stack open-sourced. Architecture, benchmarks, cost, and self-hosting notes.
Gemini 3.8 Flash Explained: Pricing Cliff, Context, Benchmarks
Gemini 3.8 Flash launched Sept 2, 2026 with a 1M context window and introductory pricing that doubles on Jan 1, 2027. A cost-per-task breakdown for builders.
GPT-6 Sol and Luna Explained: Architecture, Pricing, Benchmarks
OpenAI shipped GPT-6 Sol and Luna on Sept 22, 2026 with a 1.05M context window and prices 50% below GPT-5.6. Here is what changed and how they compare.
Claude Opus 5.5: Anthropic’s New Flagship, Benchmarked
Claude Opus 5.5 is Anthropic's Sept 2026 flagship: pricing, 1M context, verified Terminal-Bench/HLE scores, and honest gaps vs GPT-6 Astra.
Grok 4.5 Explained: Architecture, Benchmarks and Deployment (2026)
Grok 4.5 explained: xAI's 2026 flagship - MoE architecture, 500K context, Cursor-tuned agentic coding, benchmark scores, API pricing, deployment and honest limits.
Claude Opus 5 Explained: Architecture, Benchmarks and Deployment (2026)
Claude Opus 5 explained: Anthropic's 2026 frontier model - 1M-token context, effort toggle, 96% SWE-bench Verified, pricing, deployment and honest limits.
Kimi K3 Explained: Moonshot’s 2.8T Open-Weight Reasoning Model (2026)
Kimi K3 explained: Moonshot AI's 2.8T-parameter open-weight MoE - Kimi Delta Attention, 1M-token context, training, benchmarks, deployment cost and honest limits of the reasoning model in 2026.
Physical Intelligence pi0.5 Explained: The VLA Robot Foundation Model (2026)
Physical Intelligence pi0.5 explained: the ~3.3B vision-language-action robot foundation model - architecture, FAST action tokenization, training on 10+ embodiments, benchmarks, deployment and honest limits in 2026.
Mistral Large 3 Explained: Architecture & Benchmarks (2026)
Mistral Large 3 explained: the 675B/41B Apache-2.0 sparse MoE with a 262K context window - architecture, training, benchmarks, licensing, self-hosting VRAM, pricing and honest limits in 2026.