Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-09-03 · source-backed
Lily is Apache 2.0 inside pplx-garden, a small Metal inference server built for Qwen3.6-35B-A3B converted to MLX affine 4-bit, exposing a minimal OpenAI-compatible chat API with greedy decoding. The README explicitly rules out dense and smaller Qwen checkpoints, BF16, GGUF, AWQ, GPTQ, int8 and fp8, and requires Apple GPU family 10 or later (M5+) plus macOS 26. The deliberate one-model scope is the engineering choice I'd defend: a server that supports one thing well beats a matrix of half-tested paths.
Each link below shares sources, entities, or timing with this story.
Claims up to 2x faster generation, and adds fine-tuning of both MoE models on text or image datasets on Apple Silicon via MLX (release). Follow-up turns in long Qwen chats on Mac are reported up to 30x faster, MLX models now use full context size, and GLM-5.3 MLX fine-tunes ex...
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-...
Google released Gemma 4 on April 2 with four model variants: E2B, E4B, 26B MoE, and 31B Dense. The license change is the first thing worth noting. Every previous Gemma had restrictions that made lawyers nervous. Gemma 4 is Apache 2.0. Full stop. Use it in any product, any way...
AlexsJones/llmfit released v1.1.10 today, adding RamaLama runtime discovery to its MCP server, the Qwen3.8 model family and MiniMax M3 vision capability exposure (GitHub). It also merged 32 MLX benchmark results on an Apple M4 Pro, the project's first MLX entries, giving an ap...
The repo reached 9,740 stars with roughly double the next-fastest Python project's daily gain. It exposes a VAD → STT → LLM → TTS pipeline behind an OpenAI Realtime-compatible WebSocket API with every stage swappable: Parakeet TDT as default STT (Whisper, Paraformer alternativ...
The Apple Silicon inference server at 18,843 stars shipped 0.6.0 yesterday with experimental distributed serving using tensor or pipeline parallelism, capability-aware planning and memory guards (GitHub). A 225 GB MiniMax-M3 checkpoint loaded across a 128 GB and a 256 GB Mac....
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.