Blog
Articles, guides, and insights on local artificial intelligence
10 articles found per "moe"

Mistral Large 4 in open weights: a 1T MoE (52B active) for on-premise inference
Mistral released Large 4 in open weights (1.03T total, 52B active, 1M context, open-weight license): an article teaching how to read MoE specs and do the RAM, quantization and speed math for on-premise inference.

Nemotron 3.5 Bolt: NVIDIA's 74B MoE with ~9B active — what changes on your desktop
74B parameters, 9B active: NVIDIA's MoE for local agents fits 24GB of VRAM via IQ2 quantization. Real VRAM, declared benchmarks and honest limits.

MiniCPM-V 4.7: the 35B MoE that runs on a smartphone with 3 GB of RAM
MiniCPM-V 4.7 (35B total, ~3B active, released October 12, 2026 by OpenBMB and ModelBest) is the first open MoE in the series, vision-language: multiple images, video, OCR. 13+ tok/s with 3 GB of RAM on an iPhone 17 Pro Max, 17+ on Android. Official quantizations BF16, FP8, INT4, AWQ, GPTQ, GGUF; engines Ollama, llama.cpp, vLLM, SGLang. The three techniques: Lookahead Expert Scheduling, fine per-level quantization, DSpark speculative sampling.

Ling 3.1 Flash: 560B open-weight with MIT license and 1M context — what changes if you have 24GB of VRAM
inclusionAI's Ling 3.1 Flash: hybrid MoE 560B total / 25B active, native 1M-token context, FP8 checkpoint, MIT license. What it concretely means for 24GB of VRAM: quantizations, parallelism and API costs.

Aleph Alpha's Kolibri: the 78B MoE that runs (almost) locally
Aleph Alpha's Kolibri (78.1B/3.46B, Apache 2.0, FP8) is the European open-weight MoE: reasoning, tool calling and context up to 1M tokens. In FP8 it weighs ~78GB of VRAM: it runs on an RTX 5090 32GB + shared RAM or on unified-memory Macs; for practical adoption the kolibri1 architecture in llama.cpp and Ollama/LM Studio builds are still missing.

Decision models in local: JEV-35B and the 3B-active System 1
You don't need a big LLM to decide. JEV-35B (35B, 3B active) in NVFP4 on a 32GB GPU and ej (11.4MB) on CPU show how to do routing, classification and quality control without generating a single token.

Mac Studio M5 Ultra: 512GB of unified memory on the desktop
Four dies, 512GB of unified memory at 1.2 TB/s: the Mac Studio M5 Ultra runs 100-560B MoE models locally. The speed math, the cost comparison with the DGX Station, and when the Mac is the right choice.

gpt-oss: OpenAI's open reasoning model
21B total parameters but only 3.6B active, native MXFP4 weights and 128K of context: OpenAI's first open reasoning model fits in 16GB. What that means in practice.

Qwen3.8-27B vs Gemma 4 26B A4B: the new frontier at 24GB
A 27.8B dense or a MoE with 3.8B active, a 262K context against 256K and almost the same weight at Q4_K_M: two opposite ways of putting big-model quality on a 24GB consumer card. Which is faster, which goes deeper, and when to pick one over the other.

How to Read a Model's Spec Sheet: Family, Size, Context
A guide to reading a model's specs: family, total and active size, context window, quantization, benchmarks and license, with a worked example on DeepSeek V4.1 Flash.

