Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-08-29 · source-backed
78 layers where layer one is dense FFN and the other 77 are MoE, each with 256 routed experts and 1 shared expert, top-8 routing per token, plus a native 10B MTP layer (0.7B activated) built in for speculative decoding. FP8 and base variants released together on August 28; the base repo has 263 likes and 1,394 downloads. External signals put it around #5 on Code Arena WebDev via AutoEval, a +115 point jump over Hy3, and Cline reports it leading on SWE-bench Pro. Apache 2.0 on a 770B model with MTP baked in is a serious release, and the download count says almost nobody can run it. (Hugging Face)
Each link below shares sources, entities, or timing with this story.
DeepSeek released V4 Preview on April 24 with two open-weight variants: V4-Pro (1.6T total parameters, 49B activated via MoE) and V4-Flash (284B parameters, 13B activated). Both support 1M-token context windows. Both are Apache 2.0 licensed. Both are live right now on Hugging...
On July 14, llama.cpp merged native support for Tencent's Hunyuan Hy3 architecture (PR #25395), a 295B-parameter, 21B-active MoE. Any recent master build can load it now. Community GGUF quants (Q2_K, IQ2_M, Q4_K_M) from AngelSlim and others already ship on Hugging Face, and so...
Released August 28 with 78 layers, 77 of them MoE with 256 routed plus one shared expert and top-8 routing, plus a native 10B MTP layer for speculative decoding (GitHub). The attention stack uses Gated DeepSeek Sparse Attention with IndexCache for cross-layer sparse index reus...
For about a year, "run your agent locally" meant accepting a model that couldn't reliably call a tool twice in a row. That excuse is gone. Meta Superintelligence Labs published Muse Glimmer today: a 29.6B dense causal transformer, 52 layers, 6,656 hidden dim, with a ~1.8B ViT-...
The August 14 report covers January through August 2026: model repos grew from 2.43M to 2.96M, datasets from 711K to 1M, and 85.6% of models have under 200 lifetime downloads (Hugging Face). Chinese labs shipped monthly parameter ceilings of 754B to 2.78T against sub-130B for...
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...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.