Qwen3.6-27B at 4-bit: Why NVFP4 Quantization Came Down to Hopper
NVIDIA’s Qwen3.6-27B-NVFP4 compresses a 27B hybrid-attention reasoning model to 4-bit, cutting memory by roughly 2.5x while keeping benchmark gaps within 1 p...
Open Workflow Management
The Complete Gemma 4 Lineup: Five Apache 2.0 Open-Weight Models and an On-Premises Serving Guide
Released by Google DeepMind in April 2026, Gemma 4 is a multimodal open-weight family of five models spanning E2B to 31B. We break down the Apache 2.0 licens...
Running Gemma 4 26B 16x Parallel on a Single DGX Spark: NVFP4 Quantization and an Honest Cost-Efficiency Review
NVIDIA’s Gemma-4-26B-A4B-NVFP4 running 16 parallel streams on a single DGX Spark (128 GB unified memory) delivers roughly 18 tokens/s per stream and about 30...
NVIDIA Nemotron-3-Ultra-550B: LatentMoE Hybrid, 1M Context, Korean Language Support On-Premises Analysis
NVIDIA’s Nemotron-3-Ultra-550B-A55B, released under the OpenMDW-1.1 license, is a LatentMoE hybrid architecture combining Mamba-2, MoE, and Attention. A mini...
MiniMax-M3: 428B MoE Multimodal, 1M Context and MSA Deliver 9x Faster Prefill Than M2.7
MiniMaxAI’s M3 is a 428B total / 23B active parameter MoE multimodal VLM. 1M context, MiniMax Sparse Attention, SGLang/vLLM support. License is minimax-commu...
MiniMax-M2.7: 229B Open-Weight Agent Model, 113 Quantization Variants, and License Considerations
MiniMax’s M2.7 offers 229B parameters, FP8 support, and 113 quantization variants, providing a wide range of on-premises deployment paths. It claims agent te...
Kimi K2.6: 1T MoE, 32B Active, 300-Subagent Swarm Architecture Analysis
Moonshot AI’s Kimi K2.6 is a MoE model with 1T total parameters but only 32B active per token, maintaining dense 32B inference costs while supporting 256K co...
GLM-5.2: 753B MoE, 1M Context, MIT License Self-Hosting Guide
Z.ai’s GLM-5.2 handles 1M context with 2.9x FLOPs savings using DSA (Dynamic Sparse Attention). The MIT license removes barriers to on-premises deployment, a...
FastContext-1.0-4B-SFT: A 4B Subagent Model Dedicated to Repository Exploration for Coding Agents
Microsoft released FastContext-1.0-4B-SFT, a fine-tuned Qwen3-4B coding agent subagent model. It explores repositories using only three tools, READ, GLOB, an...
DiffusionGemma 26B-A4B: Google’s Experiment Generating 15-20 Tokens at Once with Discrete Text Diffusion
Google DeepMind released diffusiongemma-26B-A4B-it, a MoE-based VLM that generates text via discrete diffusion rather than autoregressive decoding. 25.2B tot...