Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-07-28 · source-backed
Korea's Upstage released a commercially usable agent-first MoE (320 routed + 1 shared experts, top-8) trained on ~12T tokens over 2M B200 GPU-hours. The architecture is deliberately serving-cheap: hybrid softmax + linear attention in a [Softmax, Linear×3]×12 pattern with NoPE and 1M context, fitting four H200s in BF16 or two quantized. CEO Kim Sung-hun pointedly contrasted this with models needing 16 B200s. Reported: 70.4 SWE-Bench Verified, 92.4 LiveCodeBench, 86.2 MMLU-Pro, 58.2 MCP-Atlas tool calling. Their Selective Weight Transfer init hit target loss in ~12B tokens versus ~22B for random. (Upstage)
Each link below shares sources, entities, or timing with this story.
This one rearranged my week. An essay published August 4 walks through Databricks' independent benchmark of coding harnesses against its own multi-million-line codebase. Pi, a harness with four built-in tools and a system prompt under 1,000 tokens, paired with Opus 4.8 at xhig...
Three moves, two days, no coordination between them. August 10–11: GitHub shipped Ollama as a BYOK provider inside Copilot for JetBrains (GitHub Changelog). Unsloth released Unsloth Desktop with a command literally named unsloth start claude, which points Claude Code and Codex...
Moonshot AI dropped Kimi K2.6 today and the numbers are hard to ignore. One trillion parameters total, 32 billion active per token across 384 experts, 256K context window, and native multimodal input. It scores 58.6 on SWE-Bench Pro versus GPT-5.4's 57.7 and Claude Opus 4.6's...
DeepSeek released V4 on April 24 and the numbers demand attention. V4-Pro is 1.6 trillion parameters total with 49 billion active, MIT-licensed, native 1M-token context. It scores 80.6% on SWE-bench Verified, putting it within 0.2 points of Claude Opus 4.6. On Terminal-Bench 2...
MAI-Code-1-Flash, a 5B-parameter coding model, is in GitHub Copilot and VS Code, and Microsoft says it beats Claude Haiku 4.5 across core coding benchmarks, +16 points on SWE-Bench Pro at 51.2% versus 35.2%, using up to 60% fewer tokens. MAI-Thinking-1, a 35B-active MoE with a...
Alibaba released Qwen3.6-27B on April 22. Dense architecture. Open weights. 77.2% on SWE-bench Verified, within 3.7 points of Claude Opus 4.6. On SkillsBench, it scores 48.2% versus its own 397B MoE predecessor's 30.0%. That's a 77% improvement with 14.8x fewer parameters. Let...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.